fix: change docker miniconda install to workspace

* feat(example): add gpt-oss-safeguard docs

* fix: add doc on reasoning_effort

.coveragerc

@@ -2,6 +2,7 @@
 source = axolotl
 omit =
     */tests/*
+    setup.py
 
 [report]
 exclude_lines =

.github/CONTRIBUTING.md

@@ -29,18 +29,13 @@ PRs are **greatly welcome**!
 2. Set up the development environment by following the instructions in the [README.md](https://github.com/axolotl-ai-cloud/axolotl/tree/main/README.md) file.
 3. Explore the codebase, run tests, and verify that everything works as expected.
 
-Please run the below to setup:
-
+Please run below to setup env
 ```bash
-git clone https://github.com/axolotl-ai-cloud/axolotl.git
-cd axolotl
+pip3 install -r requirements-dev.txt -r requirements-tests.txt
+pre-commit install
 
-uv sync --dev && uv pip install flash-attn --no-build-isolation
-source .venv/bin/activate
-
-pre-commit install  # install pre-commit hooks
-
-pytest tests/  # optional; run test suite
+# test
+pytest tests/
 ```
 
 ## How to Contribute

.github/workflows/base.yml

@@ -25,18 +25,11 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - cuda: "124"
-            cuda_version: 12.4.1
-            cudnn_version: ""
-            python_version: "3.11"
-            pytorch: 2.6.0
-            torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
-            dockerfile: "Dockerfile-base"
           - cuda: "126"
             cuda_version: 12.6.3
             cudnn_version: ""
             python_version: "3.11"
-            pytorch: 2.6.0
+            pytorch: 2.7.0
             torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
             dockerfile: "Dockerfile-base"
           - cuda: "126"
@@ -60,6 +53,20 @@ jobs:
             pytorch: 2.8.0
             torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
             dockerfile: "Dockerfile-base"
+          - cuda: "128"
+            cuda_version: 12.8.1
+            cudnn_version: ""
+            python_version: "3.11"
+            pytorch: 2.9.0
+            torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
+            dockerfile: "Dockerfile-base"
+          - cuda: "130"
+            cuda_version: 13.0.0
+            cudnn_version: ""
+            python_version: "3.11"
+            pytorch: 2.9.0
+            torch_cuda_arch_list: "9.0+PTX"
+            dockerfile: "Dockerfile-base"
 #          - cuda: "128"
 #            cuda_version: 12.8.1
 #            cudnn_version: ""
@@ -98,9 +105,7 @@ jobs:
           context: .
           file: ./docker/${{ matrix.dockerfile }}
           push: ${{ github.event_name != 'pull_request' }}
-          tags: |
-            ${{ steps.metadata.outputs.tags }}-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
-            ${{ steps.metadata.outputs.tags }}-base-uv-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
+          tags: ${{ steps.metadata.outputs.tags }}-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
           labels: ${{ steps.metadata.outputs.labels }}
           build-args: |
             CUDA_VERSION=${{ matrix.cuda_version }}
@@ -117,13 +122,6 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - cuda: "126"
-            cuda_version: 12.6.3
-            cudnn_version: ""
-            python_version: "3.11"
-            pytorch: 2.6.0
-            torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
-            dockerfile: "Dockerfile-uv-base"
           - cuda: "126"
             cuda_version: 12.6.3
             cudnn_version: ""
@@ -145,6 +143,20 @@ jobs:
             pytorch: 2.8.0
             torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
             dockerfile: "Dockerfile-uv-base"
+          - cuda: "128"
+            cuda_version: 12.8.1
+            cudnn_version: ""
+            python_version: "3.11"
+            pytorch: 2.9.0
+            torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
+            dockerfile: "Dockerfile-uv-base"
+          - cuda: "130"
+            cuda_version: 13.0.0
+            cudnn_version: ""
+            python_version: "3.11"
+            pytorch: 2.9.0
+            torch_cuda_arch_list: "9.0+PTX"
+            dockerfile: "Dockerfile-uv-base"
     steps:
       - name: Checkout
         uses: actions/checkout@v4

.github/workflows/docs.yml

@@ -20,14 +20,10 @@ jobs:
           uses: actions/setup-python@v5
           with:
             python-version: '3.11'
-        - name: Install uv
-          uses: astral-sh/setup-uv@v4
-          with:
-            version: "latest"
         - name: Install dependencies
           run: |
-            uv pip install --system jupyter quartodoc
-            uv pip install --system -e .
+            python3 -m pip install jupyter quartodoc
+            python3 -m pip install -e .
         - name: Build autodoc
           run: quartodoc build
         - name: Publish to GitHub Pages (and render)

.github/workflows/lint.yml

@@ -6,7 +6,7 @@ on:
       types: [opened, synchronize, reopened, ready_for_review]
       paths:
        - '**.py'
-       - 'pyproject.toml'
+       - 'requirements.txt'
        - '.github/workflows/*.yml'
        - "*.[q]md"
        - "examples/**/*.y[a]?ml"
@@ -23,4 +23,5 @@ jobs:
       - uses: actions/setup-python@v5
         with:
           python-version: "3.11"
+          cache: 'pip' # caching pip dependencies
       - uses: pre-commit/action@v3.0.1

.github/workflows/main.yml

@@ -18,7 +18,7 @@ jobs:
           - cuda: 126
             cuda_version: 12.6.3
             python_version: "3.11"
-            pytorch: 2.6.0
+            pytorch: 2.7.0
             axolotl_extras:
           - cuda: 126
             cuda_version: 12.6.3
@@ -68,8 +68,6 @@ jobs:
             PYTORCH_VERSION=${{ matrix.pytorch }}
             AXOLOTL_ARGS=${{ matrix.axolotl_args }}
             AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}
-            GIT_REF=${{ github.ref }}
-            GIT_SHA=${{ github.sha }}
           file: ./docker/Dockerfile
           push: ${{ github.event_name != 'pull_request' }}
           tags: |
@@ -88,7 +86,7 @@ jobs:
           - cuda: 126
             cuda_version: 12.6.3
             python_version: "3.11"
-            pytorch: 2.6.0
+            pytorch: 2.7.0
             axolotl_extras:
           - cuda: 126
             cuda_version: 12.6.3
@@ -140,8 +138,6 @@ jobs:
           build-args: |
             BASE_TAG=${{ github.ref_type == 'tag' && 'main' || github.ref_name }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
             CUDA=${{ matrix.cuda }}
-            GIT_REF=${{ github.ref }}
-            GIT_SHA=${{ github.sha }}
           file: ./docker/Dockerfile-cloud
           push: ${{ github.event_name != 'pull_request' }}
           tags: |
@@ -156,11 +152,6 @@ jobs:
     strategy:
       matrix:
         include:
-          - cuda: 126
-            cuda_version: 12.6.3
-            python_version: "3.11"
-            pytorch: 2.6.0
-            axolotl_extras:
           - cuda: 126
             cuda_version: 12.6.3
             python_version: "3.11"
@@ -207,8 +198,6 @@ jobs:
           build-args: |
             BASE_TAG=${{ github.ref_type == 'tag' && 'main' || github.ref_name }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
             CUDA=${{ matrix.cuda }}
-            GIT_REF=${{ github.ref }}
-            GIT_SHA=${{ github.sha }}
           file: ./docker/Dockerfile-cloud-no-tmux
           push: ${{ github.event_name != 'pull_request' }}
           tags: |

.github/workflows/multi-gpu-e2e.yml

@@ -4,6 +4,8 @@ on:
   pull_request:
     paths:
       - 'tests/e2e/multigpu/**.py'
+      - 'requirements.txt'
+      - 'setup.py'
       - 'pyproject.toml'
       - '.github/workflows/multi-gpu-e2e.yml'
       - 'src/axolotl/core/trainers/mixins/sequence_parallel.py'
@@ -24,13 +26,6 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - cuda: 126
-            cuda_version: 12.6.3
-            python_version: "3.11"
-            pytorch: 2.6.0
-            axolotl_extras:
-            num_gpus: 2
-            nightly_build: "true"
           - cuda: 126
             cuda_version: 12.6.3
             python_version: "3.11"
@@ -45,6 +40,13 @@ jobs:
             axolotl_extras: fbgemm-gpu
             num_gpus: 2
             nightly_build: "true"
+          - cuda: 128
+            cuda_version: 12.8.1
+            python_version: "3.11"
+            pytorch: 2.9.0
+            axolotl_extras: fbgemm-gpu
+            num_gpus: 2
+            nightly_build: "true"
     runs-on: [self-hosted, modal]
     timeout-minutes: 120
     steps:
@@ -54,17 +56,13 @@ jobs:
         uses: actions/setup-python@v5
         with:
           python-version: "3.11"
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
       - name: Install Modal
         run: |
           python -m pip install --upgrade pip
-          pip install modal==1.0.2 jinja2 protobuf
+          pip install modal==1.0.2 jinja2
       - name: Update env vars
         run: |
-          echo "BASE_TAG=${{ github.ref_name }}-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
+          echo "BASE_TAG=main-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
           echo "PYTORCH_VERSION=${{ matrix.pytorch}}" >> $GITHUB_ENV
           echo "AXOLOTL_ARGS=${{ matrix.axolotl_args}}" >> $GITHUB_ENV
           echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV
@@ -74,4 +72,4 @@ jobs:
           echo "CODECOV_TOKEN=${{ secrets.CODECOV_TOKEN }}" >> $GITHUB_ENV
       - name: Run tests job on Modal
         run: |
-          modal run -m cicd.multigpu
+          modal run cicd.multigpu

.github/workflows/nightlies.yml

@@ -12,16 +12,16 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - cuda: 126
-            cuda_version: 12.6.3
-            python_version: "3.11"
-            pytorch: 2.6.0
-            axolotl_extras:
           - cuda: 126
             cuda_version: 12.6.3
             python_version: "3.11"
             pytorch: 2.7.1
             axolotl_extras:
+          - cuda: 128
+            cuda_version: 12.8.1
+            python_version: "3.11"
+            pytorch: 2.8.0
+            axolotl_extras:
     runs-on: axolotl-gpu-runner
     steps:
       - name: Checkout
@@ -52,8 +52,6 @@ jobs:
             CUDA=${{ matrix.cuda }}
             PYTORCH_VERSION=${{ matrix.pytorch }}
             AXOLOTL_ARGS=${{ matrix.axolotl_args }}
-            GIT_REF=${{ github.ref }}
-            GIT_SHA=${{ github.sha }}
           file: ./docker/Dockerfile
           push: ${{ github.event_name != 'pull_request' }}
           tags: |
@@ -67,16 +65,16 @@ jobs:
     strategy:
       matrix:
         include:
-          - cuda: 126
-            cuda_version: 12.6.3
-            python_version: "3.11"
-            pytorch: 2.6.0
-            axolotl_extras:
           - cuda: 126
             cuda_version: 12.6.3
             python_version: "3.11"
             pytorch: 2.7.1
             axolotl_extras:
+          - cuda: 128
+            cuda_version: 12.8.1
+            python_version: "3.11"
+            pytorch: 2.8.0
+            axolotl_extras:
     runs-on: axolotl-gpu-runner
     steps:
       - name: Checkout
@@ -104,8 +102,6 @@ jobs:
           build-args: |
             BASE_TAG=${{ github.ref_name }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
             CUDA=${{ matrix.cuda }}
-            GIT_REF=${{ github.ref }}
-            GIT_SHA=${{ github.sha }}
           file: ./docker/Dockerfile-cloud
           push: ${{ github.event_name != 'pull_request' }}
           tags: |

.github/workflows/precommit-autoupdate.yml

@@ -2,7 +2,7 @@ name: Pre-commit auto-update
 
 on:
   schedule:
-    - cron: '0 0 * * 0'  # Run weekly
+    - cron: '0 0 1 * *'  # Run monthly
   workflow_dispatch:  # Manual kickoff
 
 jobs:
@@ -18,15 +18,10 @@ jobs:
         with:
           python-version: '3.11'
 
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
-
       - name: Update pre-commit hooks
         id: update
         run: |
-          uv pip install --system pre-commit
+          pip install pre-commit
           pre-commit autoupdate
           if [[ -n $(git status --porcelain) ]]; then
             echo "changes=true" >> $GITHUB_OUTPUT

.github/workflows/preview-docs.yml

@@ -40,15 +40,10 @@ jobs:
         with:
           python-version: '3.11'
 
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
-
       - name: Install dependencies
         run: |
-          uv pip install --system jupyter quartodoc
-          uv pip install --system -e .
+          python3 -m pip install jupyter quartodoc
+          python3 -m pip install -e .
 
       - name: Build autodoc
         run: quartodoc build

.github/workflows/pypi.yml

@@ -38,24 +38,23 @@ jobs:
         with:
           python-version: "3.11"
 
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
-
       - name: Install dependencies
         run: |
-          uv pip install --system wheel packaging==23.2
-          uv pip install --system --no-build-isolation -e ".[dev]"
+          pip3 install wheel packaging==23.2
+          pip3 install --no-build-isolation -e .
+          pip3 install -r requirements-dev.txt -r requirements-tests.txt
 
       - name: Extract tag name
         id: tag
-        run: echo "TAG_NAME=$(echo "$GITHUB_REF" | cut -d / -f 3)" >> "$GITHUB_OUTPUT"
+        run: echo ::set-output name=TAG_NAME::$(echo $GITHUB_REF | cut -d / -f 3)
 
-      - name: Build package
+      - name: Update version in setup.py
         run: |
-          uv pip install --system build
-          python -m build
+          sed -i -E 's/version="([0-9.]+)",/version="${{ steps.tag.outputs.TAG_NAME }}",/g' setup.py
+
+      - name: Build a source dist
+        run: |
+          python setup.py sdist
 
       - name: Publish package distributions to PyPI
         uses: pypa/gh-action-pypi-publish@release/v1

.github/workflows/tests-nightly.yml

@@ -13,6 +13,7 @@ jobs:
       - uses: actions/setup-python@v5
         with:
           python-version: "3.11"
+          cache: 'pip' # caching pip dependencies
       - uses: pre-commit/action@v3.0.1
         env:
           SKIP: no-commit-to-branch
@@ -25,7 +26,7 @@ jobs:
       max-parallel: 2
       matrix:
         python_version: ["3.11"]
-        pytorch_version: ["2.6.0", "2.7.0"]
+        pytorch_version: ["2.7.1", "2.8.0"]
     timeout-minutes: 20
 
     steps:
@@ -42,30 +43,32 @@ jobs:
         uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.python_version }}
+          cache: 'pip' # caching pip dependencies
 
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
+      - name: upgrade pip
+        run: |
+          pip3 install --upgrade pip
+          pip3 install --upgrade packaging==23.2 setuptools==75.8.0 wheel
 
       - name: Install PyTorch
         run: |
-          uv pip install --system torch==${{ matrix.pytorch_version }} torchvision
+          pip3 install torch==${{ matrix.pytorch_version }} torchvision
 
-      - name: Update pyproject.toml for nightly builds
+      - name: Update requirements.txt
         run: |
-          sed -i 's#"transformers==.*"#"transformers @ git+https://github.com/huggingface/transformers.git@main"#' pyproject.toml
-          sed -i 's#"peft==.*"#"peft @ git+https://github.com/huggingface/peft.git@main"#' pyproject.toml
-          sed -i 's#"accelerate==.*"#"accelerate @ git+https://github.com/huggingface/accelerate.git@main"#' pyproject.toml
-          sed -i 's#"trl==.*"#"trl @ git+https://github.com/huggingface/trl.git@main"#' pyproject.toml
-          sed -i 's#"datasets==.*"#"datasets @ git+https://github.com/huggingface/datasets.git@main"#' pyproject.toml
+          sed -i 's#^transformers.*#transformers @ git+https://github.com/huggingface/transformers.git@main#' requirements.txt
+          sed -i 's#^peft.*#peft @ git+https://github.com/huggingface/peft.git@main#' requirements.txt
+          sed -i 's#^accelerate.*#accelerate @ git+https://github.com/huggingface/accelerate.git@main#' requirements.txt
+          sed -i 's#^trl.*#trl @ git+https://github.com/huggingface/trl.git@main#' requirements.txt
+          sed -i 's#^datasets.*#datasets @ git+https://github.com/huggingface/datasets.git@main#' requirements.txt
 
       - name: Install dependencies
         run: |
-          uv pip show --system torch
-          uv pip install --system --no-build-isolation -e ".[dev]"
+          pip3 show torch
+          pip3 install --no-build-isolation -U -e .
           python scripts/unsloth_install.py | sh
           python scripts/cutcrossentropy_install.py | sh
+          pip3 install -r requirements-dev.txt -r requirements-tests.txt
 
       - name: Make sure PyTorch version wasn't clobbered
         run: |
@@ -81,6 +84,9 @@ jobs:
           pytest -v --durations=10 tests/patched/
           pytest -v --durations=10 tests/cli/
 
+      - name: cleanup pip cache
+        run: |
+          find "$(pip cache dir)/http-v2" -type f -mtime +14 -exec rm {} \;
 
   docker-e2e-tests:
     if: github.repository_owner == 'axolotl-ai-cloud'
@@ -96,14 +102,14 @@ jobs:
           - cuda: 126
             cuda_version: 12.6.3
             python_version: "3.11"
-            pytorch: 2.6.0
+            pytorch: 2.7.1
             num_gpus: 1
             axolotl_extras:
             nightly_build: "true"
-          - cuda: 126
-            cuda_version: 12.6.3
+          - cuda: 128
+            cuda_version: 12.8.1
             python_version: "3.11"
-            pytorch: 2.7.1
+            pytorch: 2.8.0
             num_gpus: 1
             axolotl_extras:
             nightly_build: "true"
@@ -114,16 +120,13 @@ jobs:
         uses: actions/setup-python@v5
         with:
           python-version: "3.11"
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
       - name: Install Modal
         run: |
-          uv pip install --system modal==1.0.2 jinja2
+          python -m pip install --upgrade pip
+          pip install modal==1.0.2 jinja2
       - name: Update env vars
         run: |
-          echo "BASE_TAG=main-base-uv-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
+          echo "BASE_TAG=main-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
           echo "PYTORCH_VERSION=${{ matrix.pytorch}}" >> $GITHUB_ENV
           echo "AXOLOTL_ARGS=${{ matrix.axolotl_args}}" >> $GITHUB_ENV
           echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV
@@ -133,7 +136,7 @@ jobs:
           echo "CODECOV_TOKEN=${{ secrets.CODECOV_TOKEN }}" >> $GITHUB_ENV
       - name: Run tests job on Modal
         run: |
-          modal run -m cicd.e2e_tests
+          modal run cicd.e2e_tests
   docker-e2e-multigpu-tests:
     if: github.repository_owner == 'axolotl-ai-cloud'
     # this job needs to be run on self-hosted GPU runners...
@@ -159,16 +162,13 @@ jobs:
         uses: actions/setup-python@v5
         with:
           python-version: "3.11"
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
       - name: Install Modal
         run: |
-          uv pip install --system modal==1.0.2 jinja2
+          python -m pip install --upgrade pip
+          pip install modal==1.0.2 jinja2
       - name: Update env vars
         run: |
-          echo "BASE_TAG=main-base-uv-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
+          echo "BASE_TAG=main-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
           echo "PYTORCH_VERSION=${{ matrix.pytorch}}" >> $GITHUB_ENV
           echo "AXOLOTL_ARGS=${{ matrix.axolotl_args}}" >> $GITHUB_ENV
           echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV

.github/workflows/tests.yml

@@ -7,16 +7,18 @@ on:
       - "main"
     paths:
       - '**.py'
-      - 'pyproject.toml'
+      - 'requirements.txt'
       - '.github/workflows/*.yml'
+      - 'requirements-tests.txt'
       - 'cicd/cicd.sh'
       - 'cicd/Dockerfile.jinja'
   pull_request:
       types: [opened, synchronize, reopened, ready_for_review]
       paths:
        - '**.py'
-       - 'pyproject.toml'
+       - 'requirements.txt'
        - '.github/workflows/*.yml'
+       - 'requirements-tests.txt'
        - 'cicd/cicd.sh'
        - 'cicd/Dockerfile.jinja'
   workflow_dispatch:
@@ -39,6 +41,7 @@ jobs:
       - uses: actions/setup-python@v5
         with:
           python-version: "3.11"
+          cache: 'pip' # caching pip dependencies
       - uses: pre-commit/action@v3.0.1
         env:
           SKIP: no-commit-to-branch
@@ -52,7 +55,7 @@ jobs:
       fail-fast: false
       matrix:
         python_version: ["3.11"]
-        pytorch_version: ["2.6.0", "2.7.1", "2.8.0"]
+        pytorch_version: ["2.7.1", "2.8.0", "2.9.0"]
     timeout-minutes: 20
 
     steps:
@@ -69,25 +72,24 @@ jobs:
         uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.python_version }}
+          cache: 'pip' # caching pip dependencies
 
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
+      - name: upgrade pip
+        run: |
+          pip3 install --upgrade pip
+          pip3 install --upgrade packaging==23.2 setuptools==75.8.0 wheel
 
       - name: Install PyTorch
         run: |
-          uv pip install --system torch==${{ matrix.pytorch_version }} torchvision
+          pip3 install --no-cache-dir torch==${{ matrix.pytorch_version }} torchvision
 
       - name: Install dependencies
         run: |
-          uv pip show --system torch
-          uv pip install --system wheel
-          printf "torch==${{ matrix.pytorch_version }}\n" > torch-constraints.txt
-          uv pip install --system --no-cache-dir --no-build-isolation -e ".[dev]" --constraints torch-constraints.txt
-          set -o pipefail
-          python scripts/unsloth_install.py | bash
-          python scripts/cutcrossentropy_install.py | bash
+          pip3 show torch
+          pip3 install --no-cache-dir --no-build-isolation -U -e .
+          python scripts/unsloth_install.py | sh
+          python scripts/cutcrossentropy_install.py | sh
+          pip3 install -r requirements-dev.txt -r requirements-tests.txt
 
       - name: Make sure PyTorch version wasn't clobbered
         run: |
@@ -103,10 +105,10 @@ jobs:
 
       - name: Run tests
         run: |
-          python -m pytest -v --durations=10 -n 8 --dist loadfile --cov=axolotl --cov-report=xml --ignore=tests/e2e/ --ignore=tests/patched/ --ignore=tests/cli/ --ignore=tests/monkeypatch/ tests/
-          python -m pytest -v --durations=10 -n 8 --cov=axolotl --cov-append --cov-report=xml tests/monkeypatch/
-          python -m pytest -v --durations=10 -n 8 --cov=axolotl --cov-append --cov-report=xml tests/patched/
-          python -m pytest -v --durations=10 -n 8 --cov=axolotl --cov-append --cov-report=xml tests/cli/
+          pytest -v --durations=10 -n8 --dist loadfile --ignore=tests/e2e/ --ignore=tests/patched/ --ignore=tests/cli/ --ignore=tests/monkeypatch/ tests/ --cov=axolotl --cov-report=xml
+          pytest -v --durations=10 tests/monkeypatch/ --cov=axolotl --cov-append --cov-report=xml
+          pytest -v --durations=10 tests/patched/ --cov=axolotl --cov-append --cov-report=xml
+          pytest -v --durations=10 tests/cli/ --cov=axolotl --cov-append --cov-report=xml
 
       - name: Upload coverage to Codecov
         uses: codecov/codecov-action@v5
@@ -116,6 +118,9 @@ jobs:
           flags: unittests,pytorch-${{ matrix.pytorch_version }}
           fail_ci_if_error: false
 
+      - name: cleanup pip cache
+        run: |
+          find "$(pip cache dir)/http-v2" -type f -mtime +14 -exec rm {} \;
 
   pytest-sdist:
     name: PyTest from Source Dist
@@ -125,7 +130,7 @@ jobs:
       fail-fast: false
       matrix:
         python_version: ["3.11"]
-        pytorch_version: ["2.6.0", "2.7.1", "2.8.0"]
+        pytorch_version: ["2.7.1", "2.8.0", "2.9.0"]
     timeout-minutes: 20
 
     steps:
@@ -142,26 +147,25 @@ jobs:
         uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.python_version }}
+          cache: 'pip' # caching pip dependencies
 
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
+      - name: upgrade pip
+        run: |
+          pip3 install --upgrade pip
+          pip3 install --upgrade packaging==23.2 setuptools==75.8.0 setuptools_scm build wheel psutil
 
       - name: Install PyTorch
         run: |
-          uv pip install --system torch==${{ matrix.pytorch_version }} torchvision
+          pip3 install --no-cache-dir torch==${{ matrix.pytorch_version }} torchvision
 
       - name: Install dependencies
         run: |
-          uv pip show --system torch
-          uv pip install --system wheel build setuptools_scm
-          python -m build --sdist
-          printf "torch==${{ matrix.pytorch_version }}\n" > torch-constraints.txt
-          tarball_path=$(echo dist/axolotl*.tar.gz)
-          uv pip install --no-cache-dir --no-build-isolation --system "${tarball_path}[dev]" --constraints torch-constraints.txt
+          pip3 show torch
+          python -m build --no-isolation --sdist
+          pip3 install --no-cache-dir --no-build-isolation dist/axolotl*.tar.gz
           python scripts/unsloth_install.py | sh
           python scripts/cutcrossentropy_install.py | sh
+          pip3 install -r requirements-dev.txt -r requirements-tests.txt
 
       - name: Make sure PyTorch version wasn't clobbered
         run: |
@@ -176,9 +180,13 @@ jobs:
 
       - name: Run tests
         run: |
-          python -m pytest -v --durations=10 -n 8 --dist loadfile --cov=axolotl --cov-report=xml --ignore=tests/e2e/ --ignore=tests/patched/ --ignore=tests/cli/ --ignore=tests/monkeypatch/ tests/
-          python -m pytest -v --durations=10 -n 8 --cov=axolotl --cov-append --cov-report=xml tests/monkeypatch/
-          python -m pytest -v --durations=10 -n 8 tests/cli/
+          pytest -v --durations=10 -n8 --dist loadfile --ignore=tests/e2e/ --ignore=tests/patched/ --ignore=tests/cli/ --ignore=tests/monkeypatch/ tests/ --cov=axolotl --cov-report=xml
+          pytest -v --durations=10 tests/monkeypatch/ --cov=axolotl --cov-append --cov-report=xml
+          pytest -v --durations=10 tests/cli/
+
+      - name: cleanup pip cache
+        run: |
+          find "$(pip cache dir)/http-v2" -type f -mtime +14 -exec rm {} \;
 
   gate-skip-e2e:
     needs: [pre-commit, pytest, pytest-sdist]
@@ -223,19 +231,13 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - cuda: 126
-            cuda_version: 12.6.3
+          - cuda: 128
+            cuda_version: 12.8.1
             python_version: "3.11"
-            pytorch: 2.7.1
+            pytorch: 2.8.0
             num_gpus: 1
             axolotl_extras:
-          - cuda: 126
-            cuda_version: 12.6.3
-            python_version: "3.11"
-            pytorch: 2.7.1
-            num_gpus: 1
-            axolotl_extras:
-            dockerfile: "Dockerfile.jinja"
+            dockerfile: "Dockerfile-uv.jinja"
     steps:
       - name: Checkout
         uses: actions/checkout@v4
@@ -243,17 +245,13 @@ jobs:
         uses: actions/setup-python@v5
         with:
           python-version: "3.11"
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
       - name: Install Modal
         run: |
           python -m pip install --upgrade pip
-          pip install modal==1.0.2 jinja2 protobuf
+          pip install modal==1.0.2 jinja2
       - name: Update env vars
         run: |
-          echo "BASE_TAG=${{ github.ref_name }}-base-uv-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
+          echo "BASE_TAG=main-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
           echo "PYTORCH_VERSION=${{ matrix.pytorch}}" >> $GITHUB_ENV
           echo "AXOLOTL_ARGS=${{ matrix.axolotl_args}}" >> $GITHUB_ENV
           echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV
@@ -285,15 +283,15 @@ jobs:
           - cuda: 126
             cuda_version: 12.6.3
             python_version: "3.11"
-            pytorch: 2.6.0
-            num_gpus: 1
-            axolotl_extras:
-          - cuda: 128
-            cuda_version: 12.8.1
-            python_version: "3.11"
             pytorch: 2.7.1
             num_gpus: 1
             axolotl_extras:
+#          - cuda: 128
+#            cuda_version: 12.8.1
+#            python_version: "3.11"
+#            pytorch: 2.7.1
+#            num_gpus: 1
+#            axolotl_extras:
           - cuda: 128
             cuda_version: 12.8.1
             python_version: "3.11"
@@ -301,6 +299,12 @@ jobs:
             num_gpus: 1
             gpu_type: "B200"
             axolotl_extras: fbgemm-gpu
+          - cuda: 128
+            cuda_version: 12.8.1
+            python_version: "3.11"
+            pytorch: 2.9.0
+            num_gpus: 1
+            axolotl_extras:
     steps:
       - name: Checkout
         uses: actions/checkout@v4
@@ -308,17 +312,13 @@ jobs:
         uses: actions/setup-python@v5
         with:
           python-version: "3.11"
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
       - name: Install Modal
         run: |
           python -m pip install --upgrade pip
-          pip install modal==1.0.2 jinja2 protobuf
+          pip install modal==1.0.2 jinja2
       - name: Update env vars
         run: |
-          echo "BASE_TAG=${{ github.ref_name }}-base-uv-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
+          echo "BASE_TAG=main-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
           echo "PYTORCH_VERSION=${{ matrix.pytorch}}" >> $GITHUB_ENV
           echo "AXOLOTL_ARGS=${{ matrix.axolotl_args}}" >> $GITHUB_ENV
           echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV
@@ -355,17 +355,13 @@ jobs:
         uses: actions/setup-python@v5
         with:
           python-version: "3.11"
-      - name: Install uv
-        uses: astral-sh/setup-uv@v4
-        with:
-          version: "latest"
       - name: Install Modal
         run: |
           python -m pip install --upgrade pip
-          pip install modal==1.0.2 jinja2 protobuf
+          pip install modal==1.0.2 jinja2
       - name: Update env vars
         run: |
-          echo "BASE_TAG=${{ github.ref_name }}-base-uv-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
+          echo "BASE_TAG=main-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
           echo "PYTORCH_VERSION=${{ matrix.pytorch}}" >> $GITHUB_ENV
           echo "AXOLOTL_ARGS=${{ matrix.axolotl_args}}" >> $GITHUB_ENV
           echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV

.gitignore

@@ -191,5 +191,5 @@ out/
 # vim
 *.swp
 
-# setuptools-scm generated version file
+# scm auto-versioning
 src/axolotl/_version.py

.pre-commit-config.yaml

@@ -11,13 +11,13 @@ repos:
     -   id: no-commit-to-branch
         args: ['--branch', 'main']
 -   repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.12.12
+    rev: v0.14.3
     hooks:
     -   id: ruff
         args: [--fix]
     -   id: ruff-format
 -   repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.17.1
+    rev: v1.18.2
     hooks:
     - id: mypy
       additional_dependencies:

.runpod/Dockerfile

@@ -1,8 +1,9 @@
 FROM axolotlai/axolotl-cloud:main-py3.11-cu124-2.6.0
 
 COPY .runpod/requirements.txt /requirements.txt
-RUN curl -LsSf https://astral.sh/uv/install.sh | sh && \
-    /root/.local/bin/uv pip install --system -r /requirements.txt
+RUN --mount=type=cache,target=/root/.cache/pip \
+    python3 -m pip install --upgrade pip && \
+    python3 -m pip install --upgrade -r /requirements.txt
 
 # Environment settings
 ARG BASE_VOLUME="/runpod-volume"

MANIFEST.in

@@ -1,5 +1,6 @@
-include pyproject.toml
+include requirements.txt
 include README.md
 include LICENSE
+include src/setuptools_axolotl_dynamic_dependencies.py
 include src/axolotl/utils/chat_templates/templates/*.jinja
-recursive-include src/axolotl *.py
+recursive-include axolotl *.py

README.md

@@ -65,9 +65,15 @@ Features:
 - **Flexible Dataset Handling**: Load from local, HuggingFace, and cloud (S3, Azure, GCP, OCI) datasets.
 - **Cloud Ready**: We ship [Docker images](https://hub.docker.com/u/axolotlai) and also [PyPI packages](https://pypi.org/project/axolotl/) for use on cloud platforms and local hardware.
 
+
+
 ## 🚀 Quick Start - LLM Fine-tuning in Minutes
 
-**Requirements**: NVIDIA GPU (Ampere+) or AMD GPU, Python 3.11+
+**Requirements**:
+
+- NVIDIA GPU (Ampere or newer for `bf16` and Flash Attention) or AMD GPU
+- Python 3.11
+- PyTorch ≥2.7.1
 
 ### Google Colab
 
@@ -75,35 +81,15 @@ Features:
 
 ### Installation
 
-#### Project setup (uv add)
+#### Using pip
 
 ```bash
-# Install uv
-curl -LsSf https://astral.sh/uv/install.sh | sh
-
-# Initialize or enter your project
-uv init my-project && cd my-project
-uv add axolotl
-uv pip install flash-attn --no-build-isolation
-source .venv/bin/activate
+pip3 install -U packaging==23.2 setuptools==75.8.0 wheel ninja
+pip3 install --no-build-isolation axolotl[flash-attn,deepspeed]
 
 # Download example axolotl configs, deepspeed configs
 axolotl fetch examples
-axolotl fetch deepspeed_configs  # optional
-```
-
-#### Quick try (uv pip)
-
-```bash
-# Install uv if needed
-curl -LsSf https://astral.sh/uv/install.sh | sh
-
-uv pip install axolotl
-uv pip install flash-attn --no-build-isolation
-
-# Download example axolotl configs, deepspeed configs
-axolotl fetch examples
-axolotl fetch deepspeed_configs  # optional
+axolotl fetch deepspeed_configs  # OPTIONAL
 ```
 
 #### Using Docker

cicd/Dockerfile-uv.jinja

@@ -0,0 +1,53 @@
+FROM axolotlai/axolotl-base-uv:{{ BASE_TAG }}
+
+ENV TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6 9.0+PTX"
+ENV AXOLOTL_EXTRAS="{{ AXOLOTL_EXTRAS }}"
+ENV AXOLOTL_ARGS="{{ AXOLOTL_ARGS }}"
+ENV CUDA="{{ CUDA }}"
+ENV PYTORCH_VERSION="{{ PYTORCH_VERSION }}"
+ENV GITHUB_REF="{{ GITHUB_REF }}"
+ENV GITHUB_SHA="{{ GITHUB_SHA }}"
+ENV NIGHTLY_BUILD="{{ NIGHTLY_BUILD }}"
+ENV HF_HOME="{{ HF_HOME }}"
+
+RUN apt-get update && \
+    apt-get install -y --allow-change-held-packages vim curl nano libnccl2 libnccl-dev ibverbs-providers ibverbs-utils infiniband-diags librdmacm-dev librdmacm1 rdmacm-utils slurm-wlm
+
+WORKDIR /workspace
+
+RUN git clone --depth=1 https://github.com/axolotl-ai-cloud/axolotl.git
+
+WORKDIR /workspace/axolotl
+
+RUN git fetch origin +$GITHUB_REF && \
+    git checkout FETCH_HEAD
+
+# If AXOLOTL_EXTRAS is set, append it in brackets
+RUN if [ "$NIGHTLY_BUILD" = "true" ] ; then \
+        sed -i 's#^transformers.*#transformers @ git+https://github.com/huggingface/transformers.git@main#' requirements.txt; \
+        sed -i 's#^peft.*#peft @ git+https://github.com/huggingface/peft.git@main#' requirements.txt; \
+        sed -i 's#^accelerate.*#accelerate @ git+https://github.com/huggingface/accelerate.git@main#' requirements.txt; \
+        sed -i 's#^trl.*#trl @ git+https://github.com/huggingface/trl.git@main#' requirements.txt; \
+        sed -i 's#^datasets.*#datasets @ git+https://github.com/huggingface/datasets.git@main#' requirements.txt; \
+    fi
+
+RUN uv pip install packaging==23.2 setuptools==75.8.0
+RUN uv pip install torchvision
+RUN if [ "$AXOLOTL_EXTRAS" != "" ] ; then \
+        uv pip install --no-build-isolation -e .[deepspeed,flash-attn,ring-flash-attn,optimizers,ray,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \
+    else \
+        uv pip install --no-build-isolation -e .[deepspeed,flash-attn,ring-flash-attn,optimizers,ray] $AXOLOTL_ARGS; \
+    fi
+
+RUN python scripts/unsloth_install.py --uv | sh
+RUN python scripts/cutcrossentropy_install.py --uv | sh
+
+# So we can test the Docker image
+RUN uv pip install -r requirements-dev.txt -r requirements-tests.txt
+
+# fix so that git fetch/pull from remote works
+RUN git config remote.origin.fetch "+refs/heads/*:refs/remotes/origin/*" && \
+    git config --get remote.origin.fetch
+
+# helper for huggingface-login cli
+RUN git config --global credential.helper store

cicd/Dockerfile.jinja

@@ -1,10 +1,6 @@
-FROM axolotlai/axolotl-base-uv:{{ BASE_TAG }}
+FROM axolotlai/axolotl-base:{{ BASE_TAG }}
 
-SHELL ["/bin/bash", "-euxo", "pipefail", "-c"]
-
-ARG VENV_PYTHON="/workspace/axolotl-venv/bin/python"
-
-ENV TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6 9.0+PTX"
+ENV TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
 ENV AXOLOTL_EXTRAS="{{ AXOLOTL_EXTRAS }}"
 ENV AXOLOTL_ARGS="{{ AXOLOTL_ARGS }}"
 ENV CUDA="{{ CUDA }}"
@@ -13,7 +9,7 @@ ENV GITHUB_REF="{{ GITHUB_REF }}"
 ENV GITHUB_SHA="{{ GITHUB_SHA }}"
 ENV NIGHTLY_BUILD="{{ NIGHTLY_BUILD }}"
 ENV HF_HOME="{{ HF_HOME }}"
-ENV VENV_PYTHON=$VENV_PYTHON
+ENV AXOLOTL_DATASET_NUM_PROC="8"
 
 RUN apt-get update && \
     apt-get install -y --allow-change-held-packages vim curl nano libnccl2 libnccl-dev ibverbs-providers ibverbs-utils infiniband-diags librdmacm-dev librdmacm1 rdmacm-utils slurm-wlm
@@ -29,27 +25,25 @@ RUN git fetch origin +$GITHUB_REF && \
 
 # If AXOLOTL_EXTRAS is set, append it in brackets
 RUN if [ "$NIGHTLY_BUILD" = "true" ] ; then \
-        sed -i 's#"transformers[^"]*"#"transformers @ git+https://github.com/huggingface/transformers.git@main"#' pyproject.toml; \
-        sed -i 's#"peft[^"]*"#"peft @ git+https://github.com/huggingface/peft.git@main"#' pyproject.toml; \
-        sed -i 's#"accelerate[^"]*"#"accelerate @ git+https://github.com/huggingface/accelerate.git@main"#' pyproject.toml; \
-        sed -i 's#"trl[^"]*"#"trl @ git+https://github.com/huggingface/trl.git@main"#' pyproject.toml; \
-        sed -i 's#"datasets[^"]*"#"datasets @ git+https://github.com/huggingface/datasets.git@main"#' pyproject.toml; \
+        sed -i 's#^transformers.*#transformers @ git+https://github.com/huggingface/transformers.git@main#' requirements.txt; \
+        sed -i 's#^peft.*#peft @ git+https://github.com/huggingface/peft.git@main#' requirements.txt; \
+        sed -i 's#^accelerate.*#accelerate @ git+https://github.com/huggingface/accelerate.git@main#' requirements.txt; \
+        sed -i 's#^trl.*#trl @ git+https://github.com/huggingface/trl.git@main#' requirements.txt; \
+        sed -i 's#^datasets.*#datasets @ git+https://github.com/huggingface/datasets.git@main#' requirements.txt; \
     fi
 
-RUN uv pip install --python "$VENV_PYTHON" packaging==23.2 setuptools==75.8.0 pip
+RUN pip install packaging==23.2 setuptools==75.8.0 psutil
 RUN if [ "$AXOLOTL_EXTRAS" != "" ] ; then \
-        uv pip install --python "$VENV_PYTHON" --no-build-isolation -e .[ring-flash-attn,optimizers,ray,${AXOLOTL_EXTRAS}] $AXOLOTL_ARGS; \
+        pip install --no-build-isolation -e .[deepspeed,flash-attn,ring-flash-attn,optimizers,ray,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \
     else \
-        uv pip install --python "$VENV_PYTHON" --no-build-isolation -e .[ring-flash-attn,optimizers,ray] $AXOLOTL_ARGS; \
+        pip install --no-build-isolation -e .[deepspeed,flash-attn,ring-flash-attn,optimizers,ray] $AXOLOTL_ARGS; \
     fi
 
-RUN uv pip install --python "$VENV_PYTHON" --no-build-isolation flash-attn $AXOLOTL_ARGS
-
-RUN "$VENV_PYTHON" scripts/unsloth_install.py | sh
-RUN "$VENV_PYTHON" scripts/cutcrossentropy_install.py | sh
+RUN python scripts/unsloth_install.py | sh
+RUN python scripts/cutcrossentropy_install.py | sh
 
 # So we can test the Docker image
-RUN uv pip install --python "$VENV_PYTHON" -e ".[dev]"
+RUN pip install -r requirements-dev.txt -r requirements-tests.txt
 
 # fix so that git fetch/pull from remote works
 RUN git config remote.origin.fetch "+refs/heads/*:refs/remotes/origin/*" && \

cicd/cicd.sh

@@ -4,7 +4,7 @@ set -e
 python -c "import torch; assert '$PYTORCH_VERSION' in torch.__version__"
 
 # Run unit tests with initial coverage report
-uv run pytest -v --durations=10 -n8 \
+pytest -v --durations=10 -n8 \
   --ignore=tests/e2e/ \
   --ignore=tests/patched/ \
   --ignore=tests/cli \
@@ -12,36 +12,36 @@ uv run pytest -v --durations=10 -n8 \
   --cov=axolotl
 
 # Run lora kernels tests with coverage append
-uv run pytest -v --durations=10 \
+pytest -v --durations=10 \
   /workspace/axolotl/tests/e2e/patched/lora_kernels \
   --cov=axolotl \
   --cov-append
 
 # Run patched tests excluding lora kernels with coverage append
-uv run pytest --full-trace -vvv --durations=10 \
+pytest --full-trace -vvv --durations=10 \
   --ignore=tests/e2e/patched/lora_kernels \
   /workspace/axolotl/tests/e2e/patched \
   --cov=axolotl \
   --cov-append
 
 # Run solo tests with coverage append
-uv run pytest -v --durations=10 -n1 \
+pytest -v --durations=10 -n1 \
   /workspace/axolotl/tests/e2e/solo/ \
   --cov=axolotl \
   --cov-append
 
 # Run integration tests with coverage append
-uv run pytest -v --durations=10 \
+pytest -v --durations=10 \
   /workspace/axolotl/tests/e2e/integrations/ \
   --cov=axolotl \
   --cov-append
 
-uv run pytest -v --durations=10 /workspace/axolotl/tests/cli \
+pytest -v --durations=10 /workspace/axolotl/tests/cli \
   --cov=axolotl \
   --cov-append
 
 # Run remaining e2e tests with coverage append and final report
-uv run pytest -v --durations=10 \
+pytest -v --durations=10 \
   --ignore=tests/e2e/solo/ \
   --ignore=tests/e2e/patched/ \
   --ignore=tests/e2e/multigpu/ \
@@ -52,4 +52,4 @@ uv run pytest -v --durations=10 \
   --cov-append \
   --cov-report=xml:e2e-coverage.xml
 
-uv run codecov upload-process -t $CODECOV_TOKEN -f e2e-coverage.xml -F e2e,pytorch-${PYTORCH_VERSION} || true
+codecov upload-process -t $CODECOV_TOKEN -f e2e-coverage.xml -F e2e,pytorch-${PYTORCH_VERSION} || true

cicd/multigpu.py

@@ -23,7 +23,7 @@ df_args = {
     "AXOLOTL_EXTRAS": os.environ.get("AXOLOTL_EXTRAS", ""),
     "AXOLOTL_ARGS": os.environ.get("AXOLOTL_ARGS", ""),
     "PYTORCH_VERSION": os.environ.get("PYTORCH_VERSION", "2.6.0"),
-    "BASE_TAG": os.environ.get("BASE_TAG", "main-base-uv-py3.11-cu126-2.6.0"),
+    "BASE_TAG": os.environ.get("BASE_TAG", "main-base-py3.11-cu126-2.6.0"),
     "CUDA": os.environ.get("CUDA", "126"),
     "GITHUB_REF": os.environ.get("GITHUB_REF", "refs/heads/main"),
     "GITHUB_SHA": os.environ.get("GITHUB_SHA", ""),

cicd/single_gpu.py

@@ -23,7 +23,7 @@ df_args = {
     "AXOLOTL_EXTRAS": os.environ.get("AXOLOTL_EXTRAS", ""),
     "AXOLOTL_ARGS": os.environ.get("AXOLOTL_ARGS", ""),
     "PYTORCH_VERSION": os.environ.get("PYTORCH_VERSION", "2.6.0"),
-    "BASE_TAG": os.environ.get("BASE_TAG", "main-base-uv-py3.11-cu126-2.6.0"),
+    "BASE_TAG": os.environ.get("BASE_TAG", "main-base-py3.11-cu126-2.6.0"),
     "CUDA": os.environ.get("CUDA", "126"),
     "GITHUB_REF": os.environ.get("GITHUB_REF", "refs/heads/main"),
     "GITHUB_SHA": os.environ.get("GITHUB_SHA", ""),
@@ -65,8 +65,13 @@ def run_cmd(cmd: str, run_folder: str):
     import subprocess  # nosec
 
     sp_env = os.environ.copy()
-    sp_env["AXOLOTL_DATASET_PROCESSES"] = "8"
+    sp_env["AXOLOTL_DATASET_NUM_PROC"] = "8"
 
     # Propagate errors from subprocess.
-    if exit_code := subprocess.call(cmd.split(), cwd=run_folder, env=sp_env):  # nosec
-        exit(exit_code)
+    try:
+        exit_code = subprocess.call(cmd.split(), cwd=run_folder, env=sp_env)  # nosec
+        if exit_code:
+            print(f"Command '{cmd}' failed with exit code {exit_code}")
+            return exit_code
+    except Exception as e:  # pylint: disable=broad-except
+        print(f"Command '{cmd}' failed with exception {e}")

devtools/dev_chat_template.yml

@@ -13,7 +13,7 @@ datasets:
 val_set_size: 0
 output_dir: temp_debug/axolotl_outputs/model
 dataset_prepared_path: temp_debug/axolotl_outputs/data
-dataset_processes: 1
+dataset_num_proc: 1
 
 sequence_len: 4096
 sample_packing: false

docker/Dockerfile

@@ -1,19 +1,13 @@
-ARG BASE_TAG=main-base-uv
-FROM axolotlai/axolotl-base-uv:$BASE_TAG
+ARG BASE_TAG=main-base
+FROM axolotlai/axolotl-base:$BASE_TAG
 
 ARG TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6+PTX"
 ARG AXOLOTL_EXTRAS=""
 ARG AXOLOTL_ARGS=""
 ARG CUDA="118"
 ARG PYTORCH_VERSION="2.1.2"
-ARG GIT_REF="refs/heads/main"
-ARG GIT_SHA="HEAD"
-ARG VENV_PYTHON="/workspace/axolotl-venv/bin/python"
 
 ENV PYTORCH_VERSION=$PYTORCH_VERSION
-ENV GIT_REF=$GIT_REF
-ENV GIT_SHA=$GIT_SHA
-ENV VENV_PYTHON=$VENV_PYTHON
 
 RUN apt-get update && \
     apt-get install -y --allow-change-held-packages vim curl nano libnccl2 libnccl-dev rsync s3fs && \
@@ -26,19 +20,16 @@ RUN git clone --depth=1 https://github.com/axolotl-ai-cloud/axolotl.git
 
 WORKDIR /workspace/axolotl
 
-# Ensure we are on the expected commit and break Docker cache between revisions
-RUN git fetch origin "$GIT_REF" && git checkout "$GIT_SHA"
-
 # If AXOLOTL_EXTRAS is set, append it in brackets
 RUN if [ "$AXOLOTL_EXTRAS" != "" ] ; then \
-        uv pip install --python "$VENV_PYTHON" --no-build-isolation -e .[ring-flash-attn,optimizers,ray,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \
+        pip install --no-build-isolation -e .[deepspeed,flash-attn,ring-flash-attn,optimizers,ray,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \
     else \
-        uv pip install --python "$VENV_PYTHON" --no-build-isolation -e .[ring-flash-attn,optimizers,ray] $AXOLOTL_ARGS; \
+        pip install --no-build-isolation -e .[deepspeed,flash-attn,ring-flash-attn,optimizers,ray] $AXOLOTL_ARGS; \
     fi && \
-    uv pip install --python "$VENV_PYTHON" --no-build-isolation flash-attn $AXOLOTL_ARGS && \
-    "$VENV_PYTHON" scripts/unsloth_install.py | sh && \
-    "$VENV_PYTHON" scripts/cutcrossentropy_install.py | sh && \
-    uv pip install --python "$VENV_PYTHON" pytest
+    python scripts/unsloth_install.py | sh && \
+    python scripts/cutcrossentropy_install.py | sh && \
+    pip install pytest && \
+    pip cache purge
 
 # fix so that git fetch/pull from remote works with shallow clone
 RUN git config remote.origin.fetch "+refs/heads/*:refs/remotes/origin/*" && \

docker/Dockerfile-base

@@ -5,7 +5,7 @@ ARG MAX_JOBS=4
 
 FROM nvidia/cuda:$CUDA_VERSION-cudnn$CUDNN_VERSION-devel-ubuntu$UBUNTU_VERSION AS base-builder
 
-ENV PATH="/root/miniconda3/bin:${PATH}"
+ENV PATH="/workspace/miniconda3/bin:${PATH}"
 
 ARG PYTHON_VERSION="3.10"
 ARG PYTORCH_VERSION="2.1.2"
@@ -24,29 +24,35 @@ RUN apt-get update \
     && rm -rf /var/lib/apt/lists/* \
     && wget \
     https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh \
-    && mkdir /root/.conda \
-    && bash Miniconda3-latest-Linux-x86_64.sh -b \
+    && mkdir -p /workspace/.conda \
+    && bash Miniconda3-latest-Linux-x86_64.sh -b -p /workspace/miniconda3 \
     && rm -f Miniconda3-latest-Linux-x86_64.sh \
     && conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/main \
     && conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/r \
     && conda create -n "py${PYTHON_VERSION}" python="${PYTHON_VERSION}"
 
-ENV PATH="/root/miniconda3/envs/py${PYTHON_VERSION}/bin:${PATH}"
+ENV PATH="/workspace/miniconda3/envs/py${PYTHON_VERSION}/bin:${PATH}"
 
 WORKDIR /workspace
 
-RUN python3 -m pip install --upgrade pip && pip3 install -U packaging==23.2 setuptools==75.8.0 wheel && \
+RUN python3 -m pip install --upgrade pip && pip3 install -U packaging==23.2 setuptools==75.8.0 wheel psutil && \
     python3 -m pip install --no-cache-dir -U torch==${PYTORCH_VERSION}+cu${CUDA} torchvision --extra-index-url https://download.pytorch.org/whl/cu$CUDA && \
-    CAUSAL_CONV1D_FORCE_CXX11_ABI=TRUE CAUSAL_CONV1D_FORCE_BUILD=TRUE python3 -m pip install --no-cache-dir causal_conv1d==1.5.2 && \
-    python3 -m pip install --no-cache-dir "mamba_ssm @ git+https://github.com/state-spaces/mamba.git@main" && \
     python3 -m pip cache purge
 
+RUN if [ "$CUDA" != "130" ] ; then \
+        CAUSAL_CONV1D_FORCE_CXX11_ABI=TRUE CAUSAL_CONV1D_FORCE_BUILD=TRUE python3 -m pip install --no-cache-dir "causal_conv1d @ git+https://github.com/Dao-AILab/causal-conv1d.git@v1.5.4"; \
+        python3 -m pip install --no-cache-dir "mamba_ssm @ git+https://github.com/state-spaces/mamba.git@main"; \
+        python3 -m pip cache purge; \
+    fi
+
 RUN git lfs install --skip-repo && \
     pip3 install awscli && \
     # The base image ships with `pydantic==1.8.2` which is not working
     pip3 install -U --no-cache-dir pydantic==1.10.10 && \
     pip3 cache purge
 
-RUN if [ "$PYTORCH_VERSION" = "2.6.0" ] && [ "$CUDA" = "124" ] ; then \
-        FLASH_ATTENTION_FORCE_BUILD="TRUE" uv pip install --no-build-isolation flash-attn==2.8.0.post2; \
+RUN if [ "$PYTORCH_VERSION" = "2.9.0" ] && [ "$CUDA" = "128" ] ; then \
+        wget https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.4.17/flash_attn-2.8.3+cu128torch2.9-cp311-cp311-linux_x86_64.whl; \
+        pip3 install --no-cache-dir flash_attn-2.8.3+cu128torch2.9-cp311-cp311-linux_x86_64.whl; \
+        rm flash_attn-2.8.3+cu128torch2.9-cp311-cp311-linux_x86_64.whl; \
     fi

docker/Dockerfile-base-next

@@ -5,7 +5,7 @@ ARG MAX_JOBS=4
 
 FROM nvidia/cuda:$CUDA_VERSION-cudnn$CUDNN_VERSION-devel-ubuntu$UBUNTU_VERSION AS base-builder
 
-ENV PATH="/root/miniconda3/bin:${PATH}"
+ENV PATH="/workspace/miniconda3/bin:${PATH}"
 
 ARG PYTHON_VERSION="3.11"
 ARG PYTORCH_VERSION="next"
@@ -19,12 +19,12 @@ RUN apt-get update \
     && apt-get install -y wget git build-essential ninja-build git-lfs libaio-dev pkg-config && rm -rf /var/lib/apt/lists/* \
     && wget \
     https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh \
-    && mkdir /root/.conda \
-    && bash Miniconda3-latest-Linux-x86_64.sh -b \
+    && mkdir -p /workspace/.conda \
+    && bash Miniconda3-latest-Linux-x86_64.sh -b -p /workspace/miniconda3 \
     && rm -f Miniconda3-latest-Linux-x86_64.sh \
     && conda create -n "py${PYTHON_VERSION}" python="${PYTHON_VERSION}"
 
-ENV PATH="/root/miniconda3/envs/py${PYTHON_VERSION}/bin:${PATH}"
+ENV PATH="/workspace/miniconda3/envs/py${PYTHON_VERSION}/bin:${PATH}"
 
 WORKDIR /workspace
 

docker/Dockerfile-base-nightly

@@ -5,7 +5,7 @@ ARG MAX_JOBS=4
 
 FROM nvidia/cuda:$CUDA_VERSION-cudnn$CUDNN_VERSION-devel-ubuntu$UBUNTU_VERSION AS base-builder
 
-ENV PATH="/root/miniconda3/bin:${PATH}"
+ENV PATH="/workspace/miniconda3/bin:${PATH}"
 
 ARG PYTHON_VERSION="3.11"
 ARG PYTORCH_VERSION="nightly"
@@ -19,14 +19,14 @@ RUN apt-get update \
     && apt-get install -y wget git build-essential ninja-build git-lfs libaio-dev pkg-config && rm -rf /var/lib/apt/lists/* \
     && wget \
     https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh \
-    && mkdir /root/.conda \
-    && bash Miniconda3-latest-Linux-x86_64.sh -b \
+    && mkdir -p /workspace/.conda \
+    && bash Miniconda3-latest-Linux-x86_64.sh -b -p /workspace/miniconda3 \
     && rm -f Miniconda3-latest-Linux-x86_64.sh \
     && conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/main \
     && conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/r \
     && conda create -n "py${PYTHON_VERSION}" python="${PYTHON_VERSION}"
 
-ENV PATH="/root/miniconda3/envs/py${PYTHON_VERSION}/bin:${PATH}"
+ENV PATH="/workspace/miniconda3/envs/py${PYTHON_VERSION}/bin:${PATH}"
 
 WORKDIR /workspace
 

docker/Dockerfile-cloud

@@ -12,8 +12,8 @@ EXPOSE 22
 COPY scripts/cloud-entrypoint.sh /root/cloud-entrypoint.sh
 COPY scripts/motd /etc/motd
 
-RUN uv pip install --python "$VENV_PYTHON" jupyterlab notebook ipywidgets && \
-    "$VENV_PYTHON" -m jupyter lab clean
+RUN pip install jupyterlab notebook ipywidgets && \
+    jupyter lab clean
 RUN apt update && \
     apt install --yes --no-install-recommends openssh-server tmux iproute2 nvtop && \
     rm -rf /var/cache/apt/archives && \

docker/Dockerfile-cloud-no-tmux

@@ -12,8 +12,8 @@ EXPOSE 22
 COPY scripts/cloud-entrypoint.sh /root/cloud-entrypoint.sh
 COPY scripts/motd /etc/motd
 
-RUN uv pip install --python "$VENV_PYTHON" jupyterlab notebook ipywidgets && \
-    "$VENV_PYTHON" -m jupyter lab clean
+RUN pip install jupyterlab notebook ipywidgets && \
+    jupyter lab clean
 RUN apt update && \
     apt install --yes --no-install-recommends openssh-server tmux iproute2 nvtop ibverbs-providers ibverbs-utils infiniband-diags librdmacm-dev librdmacm1 rdmacm-utils slurm-wlm && \
     rm -rf /var/cache/apt/archives && \

docker/Dockerfile-tests

@@ -24,14 +24,13 @@ RUN git fetch origin +$GITHUB_REF && \
 
 # If AXOLOTL_EXTRAS is set, append it in brackets
 RUN if [ "$AXOLOTL_EXTRAS" != "" ] ; then \
-        uv pip install --no-build-isolation -e .[deepspeed,mamba-ssm,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \
+        pip install --no-build-isolation -e .[deepspeed,flash-attn,mamba-ssm,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \
     else \
-        uv pip install --no-build-isolation -e .[deepspeed,mamba-ssm] $AXOLOTL_ARGS; \
-    fi && \
-    uv pip install --no-build-isolation flash-attn $AXOLOTL_ARGS
+        pip install --no-build-isolation -e .[deepspeed,flash-attn,mamba-ssm] $AXOLOTL_ARGS; \
+    fi
 
 # So we can test the Docker image
-RUN uv pip install pytest
+RUN pip install pytest
 
 # fix so that git fetch/pull from remote works
 RUN git config remote.origin.fetch "+refs/heads/*:refs/remotes/origin/*" && \

docker/Dockerfile-uv-base

@@ -13,7 +13,6 @@ ARG TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6 9.0+PTX"
 ENV PYTHON_VERSION=$PYTHON_VERSION
 ENV TORCH_CUDA_ARCH_LIST=$TORCH_CUDA_ARCH_LIST
 ENV UV_TORCH_BACKEND="cu${CUDA}"
-ENV VENV_PYTHON=/workspace/axolotl-venv/bin/python
 
 RUN apt-get update \
     && apt-get install -y wget git build-essential ninja-build git-lfs libaio-dev pkg-config curl && rm -rf /var/lib/apt/lists/* \
@@ -30,8 +29,14 @@ RUN uv venv --no-project --relocatable axolotl-venv
 
 ENV PATH="/workspace/axolotl-venv/bin:${PATH}"
 
-RUN uv pip install --python "$VENV_PYTHON" packaging setuptools wheel psutil protobuf grpclib \
-    && uv pip install --python "$VENV_PYTHON" torch==${PYTORCH_VERSION} \
-    && uv pip install --python "$VENV_PYTHON" --no-build-isolation "causal_conv1d @ git+https://github.com/Dao-AILab/causal-conv1d.git@main" \
-    && uv pip install --python "$VENV_PYTHON" "mamba_ssm @ git+https://github.com/state-spaces/mamba.git@main" \
-    && uv pip install --python "$VENV_PYTHON" awscli pydantic
+RUN uv pip install packaging setuptools wheel psutil \
+    && uv pip install torch==${PYTORCH_VERSION} torchvision \
+    && uv pip install --no-build-isolation "causal_conv1d @ git+https://github.com/Dao-AILab/causal-conv1d.git@main" \
+    && uv pip install "mamba_ssm @ git+https://github.com/state-spaces/mamba.git@main" \
+    && uv pip install awscli pydantic
+
+RUN if [ "$PYTORCH_VERSION" = "2.9.0" ] && [ "$CUDA" = "128" ] ; then \
+        wget https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.4.17/flash_attn-2.8.3+cu128torch2.9-cp311-cp311-linux_x86_64.whl; \
+        uv pip install --no-cache-dir flash_attn-2.8.3+cu128torch2.9-cp311-cp311-linux_x86_64.whl; \
+        rm flash_attn-2.8.3+cu128torch2.9-cp311-cp311-linux_x86_64.whl; \
+    fi

docs/debugging.qmd

@@ -29,7 +29,7 @@ While debugging it's helpful to simplify your test scenario as much as possible.
 1. **Make sure you are using the latest version of axolotl**:  This project changes often and bugs get fixed fast.  Check your git branch and make sure you have pulled the latest changes from `main`.
 1. **Eliminate concurrency**: Restrict the number of processes to 1 for both training and data preprocessing:
     - Set `CUDA_VISIBLE_DEVICES` to a single GPU, ex: `export CUDA_VISIBLE_DEVICES=0`.
-    - Set `dataset_processes: 1` in your axolotl config or run the training command with `--dataset_processes=1`.
+    - Set `dataset_num_proc: 1` in your axolotl config or run the training command with `--dataset_num_proc=1`.
 2. **Use a small dataset**: Construct or use a small dataset from HF Hub. When using a small dataset, you will often have to make sure `sample_packing: False` and `eval_sample_packing: False` to avoid errors.  If you are in a pinch and don't have time to construct a small dataset but want to use from the HF Hub, you can shard the data (this will still tokenize the entire dataset, but will only use a fraction of the data for training.  For example, to shard the dataset into 20 pieces, add the following to your axolotl config):
 
     ```yaml
@@ -72,8 +72,8 @@ datasets:
 Make sure you have an [editable install](https://setuptools.pypa.io/en/latest/userguide/development_mode.html) of Axolotl, which ensures that changes you make to the code are reflected at runtime.  Run the following commands from the root of this project:
 
 ```bash
-uv sync --extra deepspeed
-uv pip install flash-attn --no-build-isolation
+pip3 install packaging
+pip3 install --no-build-isolation -e '.[flash-attn,deepspeed]'
 ```
 
 #### Remote Hosts
@@ -101,7 +101,7 @@ For example, to mimic the command `cd devtools && CUDA_VISIBLE_DEVICES=0 acceler
                 "-m", "axolotl.cli.train", "dev_chat_template.yml",
                 // The flags below simplify debugging by overriding the axolotl config
                 // with the debugging tips above.  Modify as needed.
-                "--dataset_processes=1",      // limits data preprocessing to one process
+                "--dataset_num_proc=1",      // limits data preprocessing to one process
                 "--max_steps=1",              // limits training to just one step
                 "--batch_size=1",             // minimizes batch size
                 "--micro_batch_size=1",       // minimizes batch size
@@ -213,8 +213,8 @@ docker run --privileged --gpus '"all"' --shm-size 10g --rm -it --name axolotl --
 You will now be in the container.  Next, perform an editable install of Axolotl:
 
 ```bash
-uv sync --extra deepspeed
-uv pip install flash-attn --no-build-isolation
+pip3 install packaging
+pip3 install --no-build-isolation -e '.[flash-attn,deepspeed]'
 ```
 
 ### Attach To Container

docs/faq.qmd

@@ -63,6 +63,14 @@ description: Frequently asked questions
 
 > A: There seems to be a wheel issue with FA2 2.8.0 on CUDA 12.4. Try CUDA 12.6 instead or downgrade to FA2 2.7.4. Please refer to the upstream issue: https://github.com/Dao-AILab/flash-attention/issues/1717.
 
+**Q: Can we mix text and text+image datasets for VLM training?**
+
+> A: Yes, you can for newer VLM arch. The ones that would not work are LLaVA / Pixtral arch. If you notice one not working, please let us know!
+
+**Q: Why is `memory/max_*` different from `nvidia-smi`?**
+
+> A: We use `torch` APIs to retrieve this information. You can see https://docs.pytorch.org/docs/stable/notes/cuda.html#cuda-memory-management for more information.
+
 ### Chat templates
 
 **Q: `jinja2.exceptions.UndefinedError: 'dict object' has no attribute 'content' / 'role' / ____`**

docs/installation.qmd

@@ -29,40 +29,19 @@ Follow the instructions at: [https://pytorch.org/get-started/locally/](https://p
 For Blackwell GPUs, please use Pytorch 2.7.0 and CUDA 12.8.
 :::
 
-### uv Installation (Recommended) {#sec-uv-quick}
+### PyPI Installation (Recommended) {#sec-pypi}
 
 ```{.bash}
-# Install uv if not already installed
-curl -LsSf https://astral.sh/uv/install.sh | sh
-
-# Add Axolotl to a project (recommended)
-uv init my-project && cd my-project
-uv add axolotl
-uv pip install flash-attn --no-build-isolation
-source .venv/bin/activate
-```
-
-For a quick one-off install without creating a project:
-
-```{.bash}
-uv pip install axolotl
-uv pip install flash-attn --no-build-isolation
-```
-
-### pip Installation {#sec-pypi}
-
-```{.bash}
-pip install --no-build-isolation axolotl[deepspeed]
-pip install --no-build-isolation flash-attn
+pip3 install -U packaging setuptools wheel ninja
+pip3 install --no-build-isolation axolotl[flash-attn,deepspeed]
 ```
 
 We use `--no-build-isolation` in order to detect the installed PyTorch version (if
 installed) in order not to clobber it, and so that we set the correct version of
 dependencies that are specific to the PyTorch version or other installed
-co-dependencies. Flash Attention is resolved separately so it can be built against
-the environment configured by the previous step.
+co-dependencies.
 
-### Advanced uv Installation {#sec-uv}
+### uv Installation {#sec-uv}
 
 uv is a fast, reliable Python package installer and resolver built in Rust. It offers significant performance improvements over pip and provides better dependency resolution, making it an excellent choice for complex environments.
 
@@ -83,38 +62,28 @@ source .venv/bin/activate
 Install PyTorch
 - PyTorch 2.6.0 recommended
 ```{.bash}
+uv pip install packaging setuptools wheel
 uv pip install torch==2.6.0
 uv pip install awscli pydantic
 ```
 
 Install axolotl from PyPi
 ```{.bash}
-uv pip install --no-build-isolation axolotl[deepspeed]
-# optionally install with vLLM if you're using torch==2.6.0 and want to train w/ GRPO
-# uv pip install --no-build-isolation axolotl[deepspeed,vllm]
+uv pip install --no-build-isolation axolotl[deepspeed,flash-attn]
 
-uv pip install flash-attn --no-build-isolation
+# optionally install with vLLM if you're using torch==2.6.0 and want to train w/ GRPO
+uv pip install --no-build-isolation axolotl[deepspeed,flash-attn,vllm]
 ```
 
 ### Edge/Development Build {#sec-edge-build}
 
 For the latest features between releases:
 
-#### Using uv (recommended)
 ```{.bash}
 git clone https://github.com/axolotl-ai-cloud/axolotl.git
 cd axolotl
-curl -LsSf https://astral.sh/uv/install.sh | sh  # If not already installed
-uv sync
-uv pip install flash-attn --no-build-isolation
-```
-
-#### Using pip
-```{.bash}
-git clone https://github.com/axolotl-ai-cloud/axolotl.git
-cd axolotl
-pip install --no-build-isolation -e '.[deepspeed]'
-pip install --no-build-isolation flash-attn
+pip3 install -U packaging setuptools wheel ninja
+pip3 install --no-build-isolation -e '.[flash-attn,deepspeed]'
 ```
 
 ### Docker {#sec-docker}
@@ -172,7 +141,7 @@ For providers supporting Docker:
 ### macOS {#sec-macos}
 
 ```{.bash}
-uv pip install --no-build-isolation -e '.'
+pip3 install --no-build-isolation -e '.'
 ```
 
 See @sec-troubleshooting for Mac-specific issues.
@@ -190,15 +159,10 @@ We recommend using WSL2 (Windows Subsystem for Linux) or Docker.
 1. Install Python ≥3.11
 2. Install PyTorch: https://pytorch.org/get-started/locally/
 3. Install Axolotl:
-```{.bash}
-# Option A: add Axolotl to the environment
-uv add axolotl
-uv pip install flash-attn --no-build-isolation
-
-# Option B: quick install
-uv pip install axolotl
-uv pip install flash-attn --no-build-isolation
-```
+   ```{.bash}
+   pip3 install -U packaging setuptools wheel ninja
+   pip3 install --no-build-isolation -e '.[flash-attn,deepspeed]'
+   ```
 4. (Optional) Login to Hugging Face:
    ```{.bash}
    huggingface-cli login

docs/lr_groups.qmd

@@ -27,3 +27,9 @@ learning_rate: 2e-5
 In this example, we have a default learning rate of 2e-5 across the entire model, but we have a separate learning rate
 of 1e-6 for all the self attention `o_proj` modules across all layers, and a learning are of 1e-5 to the 3rd layer's
 self attention `q_proj` module.
+
+::: {.callout-note}
+
+We currently only support varying `lr` for now. If you're interested in adding support for others (`weight_decay`), we welcome PRs. See https://github.com/axolotl-ai-cloud/axolotl/blob/613bcf90e58f3ab81d3827e7fc572319908db9fb/src/axolotl/core/trainers/mixins/optimizer.py#L17
+
+:::

docs/multi-gpu.qmd

@@ -88,6 +88,7 @@ fsdp_sync_module_states | **REMOVED**
 fsdp_cpu_ram_efficient_loading | cpu_ram_efficient_loading
 fsdp_state_dict_type | state_dict_type
 fsdp_use_orig_params | **REMOVED**
+fsdp_activation_checkpointing | activation_checkpointing
 
 For more details, please see the migration guide in the [torchtitan repo](https://github.com/pytorch/torchtitan/blob/main/docs/fsdp.md). In Axolotl,
 if you were using the following FSDP1 config:

docs/multimodal.qmd

@@ -56,10 +56,14 @@ image_resize_algorithm: bilinear
 
 Please see [examples](https://github.com/axolotl-ai/axolotl/tree/main/examples) folder for full configs.
 
-::: {.callout-warning}
+::: {.callout-tip}
 Some of our chat_templates have been extended to support broader dataset types. This should not break any existing configs.
 :::
 
+::: {.callout-note}
+As of now, we do not truncate nor drop samples based on `sequence_len` as each arch has different ways to process non-text tokens. We are looking for help on this.
+:::
+
 ### Mllama {#sec-mllama}
 
 ```yaml
@@ -95,7 +99,7 @@ chat_template: llava
 ### Mistral-Small-3.1 {#sec-mistral-small-31}
 
 ::: {.callout-tip}
-Please make sure to install vision lib via `uv pip install 'mistral-common[opencv]==1.8.5'`
+Please make sure to install vision lib via `pip install 'mistral-common[opencv]==1.8.5'`
 :::
 
 ```yaml
@@ -105,7 +109,7 @@ base_model: mistralai/Mistral-Small-3.1-24B-Instruct-2503
 ### Magistral-Small-2509 {#sec-magistral-small-2509}
 
 ::: {.callout-tip}
-Please make sure to install vision lib via `uv pip install 'mistral-common[opencv]==1.8.5'`
+Please make sure to install vision lib via `pip install 'mistral-common[opencv]==1.8.5'`
 :::
 
 ```yaml
@@ -115,7 +119,7 @@ base_model: mistralai/Magistral-Small-2509
 ### Voxtral {#sec-voxtral}
 
 ::: {.callout-tip}
-Please make sure to install audio lib via `uv pip install librosa==0.11.0 'mistral_common[audio]==1.8.3'`
+Please make sure to install audio lib via `pip3 install librosa==0.11.0 'mistral_common[audio]==1.8.3'`
 :::
 
 ```yaml
@@ -143,7 +147,7 @@ The model's initial loss and grad norm will be very high. We suspect this to be
 :::
 
 ::: {.callout-tip}
-Please make sure to install `timm` via `uv pip install timm==1.0.17`
+Please make sure to install `timm` via `pip3 install timm==1.0.17`
 :::
 
 ```yaml
@@ -168,10 +172,18 @@ base_model: Qwen/Qwen2.5-VL-7B-Instruct
 chat_template: qwen2_vl  # same as qwen2-vl
 ```
 
+### Qwen3-VL {#sec-qwen3-vl}
+
+```yaml
+base_model: Qwen/Qwen3-VL-4B-Instruct
+
+chat_template: qwen2_vl  # same as qwen2-vl
+```
+
 ### SmolVLM2 {#sec-smolvlm2}
 
 ::: {.callout-tip}
-Please make sure to install `num2words` via `uv pip install num2words==0.5.14`
+Please make sure to install `num2words` via `pip3 install num2words==0.5.14`
 :::
 
 ```yaml
@@ -181,7 +193,7 @@ base_model: HuggingFaceTB/SmolVLM2-500M-Video-Instruct
 ### LFM2-VL {#sec-lfm2-vl}
 
 ::: {.callout-warning}
-Please uninstall `causal-conv1d` via `uv pip uninstall -y causal-conv1d`
+Please uninstall `causal-conv1d` via `pip3 uninstall -y causal-conv1d`
 :::
 
 ```yaml
@@ -222,7 +234,7 @@ For audio loading, you can use the following keys within `content` alongside `"t
 
 ::: {.callout-tip}
 
-You may need to install `librosa` via `uv pip install librosa==0.11.0`.
+You may need to install `librosa` via `pip3 install librosa==0.11.0`.
 
 :::
 

docs/rlhf.qmd

@@ -219,6 +219,21 @@ DPO supports the following types with the following dataset format:
 }
 ```
 
+#### chat_template.argilla_chat
+
+```json
+{
+    "chosen": [
+        {"role": "user", "content": "..."},
+        {"role": "assistant", "content": "..."}
+    ],
+    "rejected": [
+        {"role": "user", "content": "..."},
+        {"role": "assistant", "content": "..."}
+    ]
+}
+```
+
 #### chat_template.default
 
 ```yaml

docs/sequence_parallelism.qmd

@@ -49,9 +49,9 @@ When sequence parallelism is enabled:
 To use sequence parallelism, you need:
 
 - Multiple GPUs (at least 2)
-- The `ring-flash-attn` package. Install with either `uv sync --extra ring-flash-attn`
-  (from a cloned repository) or `uv pip install ring-flash-attn>=0.1.4`.
-- Flash Attention installed separately with `uv pip install flash-attn --no-build-isolation`.
+- The `ring-flash-attn` package. Install with:
+  - `pip install axolotl[ring-flash-attn]` (preferred)
+  - `pip install ring-flash-attn>=0.1.4`
 
 ## Limitations
 

examples/LiquidAI/README.md

@@ -6,20 +6,17 @@ LFM2 features a new hybrid Liquid architecture with multiplicative gates, short-
 
 This guide shows how to fine-tune both the LFM2 and LFM2-VL models with Axolotl.
 
+Thanks to the team at LiquidAI for giving us early access to prepare for these releases.
+
 ## Getting Started
 
 1.  Install Axolotl following the [installation guide](https://docs.axolotl.ai/docs/installation.html).
 
     Here is an example of how to install from pip:
     ```bash
-    # Ensure you have a compatible version of PyTorch installed
-    # Option A: manage dependencies in your project
-    uv add 'axolotl>=0.12.0'
-    uv pip install flash-attn --no-build-isolation
-
-    # Option B: quick install
-    uv pip install 'axolotl>=0.12.0'
-    uv pip install flash-attn --no-build-isolation
+    # Ensure you have a compatible version of Pytorch installed
+    pip3 install packaging setuptools wheel ninja
+    pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
     ```
 
 2.  Run one of the finetuning examples below.
@@ -36,11 +33,19 @@ This guide shows how to fine-tune both the LFM2 and LFM2-VL models with Axolotl.
     axolotl train examples/LiquidAI/lfm2-vl-lora.yaml
     ```
 
+    **LFM2-MoE**
+    ```bash
+    pip install git+https://github.com/huggingface/transformers.git@0c9a72e4576fe4c84077f066e585129c97bfd4e6
+
+    # LoRA SFT (1x48GB @ 16.2GiB)
+    axolotl train examples/LiquidAI/lfm2-8b-a1b-lora.yaml
+    ```
+
 ### TIPS
 
 - **Installation Error**: If you encounter `ImportError: ... undefined symbol ...` or `ModuleNotFoundError: No module named 'causal_conv1d_cuda'`, the `causal-conv1d` package may have been installed incorrectly. Try uninstalling it:
   ```bash
-  uv pip uninstall -y causal-conv1d
+  pip uninstall -y causal-conv1d
   ```
 
 - **Dataset Loading**: Read more on how to load your own dataset in our [documentation](https://docs.axolotl.ai/docs/dataset_loading.html).
@@ -50,14 +55,13 @@ This guide shows how to fine-tune both the LFM2 and LFM2-VL models with Axolotl.
 
 ## Optimization Guides
 
-- [Multi-GPU Training](https://docs.axolotl.ai/docs/multi-gpu.html)
-- [LoRA Optimizations](https://docs.axolotl.ai/docs/lora_optims.html)
-- [Multi-Node Training](https://docs.axolotl.ai/docs/multi-node.html)
+- [Optimizations Guide](https://docs.axolotl.ai/docs/optimizations.html)
 
 ## Related Resources
 
 - [LFM2 Blog](https://www.liquid.ai/blog/liquid-foundation-models-v2-our-second-series-of-generative-ai-models)
 - [LFM2-VL Blog](https://www.liquid.ai/blog/lfm2-vl-efficient-vision-language-models)
+- [LFM2-MoE Blog](https://www.liquid.ai/blog/lfm2-8b-a1b-an-efficient-on-device-mixture-of-experts)
 - [Axolotl Docs](https://docs.axolotl.ai)
 - [Axolotl GitHub](https://github.com/axolotl-ai-cloud/axolotl)
 - [Axolotl Discord](https://discord.gg/7m9sfhzaf3)

examples/LiquidAI/lfm2-350m-fft.yaml

@@ -1,6 +1,7 @@
 base_model: LiquidAI/LFM2-350M
 
-chunked_cross_entropy: true
+plugins:
+  - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
 
 eot_tokens:
   - "<|im_end|>"

examples/LiquidAI/lfm2-8b-a1b-lora.yaml

@@ -0,0 +1,59 @@
+base_model: LiquidAI/LFM2-8B-A1B
+
+plugins:
+  - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
+
+load_in_8bit: true
+
+eot_tokens:
+  - "<|im_end|>"
+datasets:
+  - path: mlabonne/FineTome-100k
+    type: chat_template
+    split: train[:20%]
+    field_messages: conversations
+    message_field_role: from
+    message_field_content: value
+dataset_prepared_path: last_run_prepared
+val_set_size: 0.05
+output_dir: ./outputs/out
+
+sequence_len: 4096
+sample_packing: true
+
+adapter: lora
+lora_model_dir:
+
+lora_r: 32
+lora_alpha: 16
+lora_dropout: 0.05
+lora_target_modules: 'model.layers.[\d]+.(mlp|cross_attn|self_attn).(up|down|gate|q|k|v|o)_proj'
+
+wandb_project:
+wandb_entity:
+wandb_watch:
+wandb_name:
+wandb_log_model:
+
+gradient_accumulation_steps: 2
+micro_batch_size: 4
+num_epochs: 1
+optimizer: adamw_torch_fused
+lr_scheduler: cosine
+learning_rate: 5e-5
+
+bf16: true
+tf32: true
+
+gradient_checkpointing: true
+resume_from_checkpoint:
+logging_steps: 1
+flash_attention: true
+
+warmup_ratio: 0.1
+evals_per_epoch: 2
+saves_per_epoch: 1
+
+weight_decay: 0.0
+
+# save_first_step: true  # uncomment this to validate checkpoint saving works with your config

examples/LiquidAI/lfm2-vl-lora.yaml

@@ -3,6 +3,9 @@ trust_remote_code: true
 model_type: AutoModelForImageTextToText
 processor_type: AutoProcessor
 
+plugins:
+  - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
+
 # these 3 lines are needed for now to handle vision chat templates w images
 skip_prepare_dataset: true
 remove_unused_columns: false

examples/apertus/README.md

@@ -15,8 +15,8 @@ This guide shows how to fine-tune it with Axolotl with multi-turn conversations
 git clone https://github.com/axolotl-ai-cloud/axolotl.git
 cd axolotl
 
-uv sync
-uv pip install flash-attn --no-build-isolation
+pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja
+pip3 install --no-build-isolation -e '.[flash-attn]'
 
 # Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy
 python scripts/cutcrossentropy_install.py | sh
@@ -31,7 +31,7 @@ python scripts/cutcrossentropy_install.py | sh
 # For those using our Docker image, use the below path.
 export CUDA_HOME=/usr/local/cuda
 
-uv pip install git+https://github.com/nickjbrowning/XIELU@59d6031 --no-build-isolation --no-deps
+pip3 install git+https://github.com/nickjbrowning/XIELU@59d6031 --no-build-isolation --no-deps
 ```
 
 For any installation errors, see [XIELU Installation Issues](#xielu-installation-issues)
@@ -67,7 +67,7 @@ If those didn't help, please try the below solutions:
 1. Pass env for CMAKE and try install again:
 
     ```bash
-    Python_EXECUTABLE=$(which python) uv pip install git+https://github.com/nickjbrowning/XIELU@59d6031 --no-build-isolation --no-deps
+    Python_EXECUTABLE=$(which python) pip3 install git+https://github.com/nickjbrowning/XIELU@59d6031 --no-build-isolation --no-deps
     ```
 
 2. Git clone the repo and manually hardcode python path:
@@ -92,7 +92,7 @@ If those didn't help, please try the below solutions:
     ```
 
     ```bash
-    uv pip install . --no-build-isolation --no-deps
+    pip3 install . --no-build-isolation --no-deps
     ```
 
 ## Optimization Guides

examples/arcee/README.md

@@ -17,8 +17,8 @@ Thanks to the team at Arcee.ai for using Axolotl in supervised fine-tuning the A
 git clone https://github.com/axolotl-ai-cloud/axolotl.git
 cd axolotl
 
-uv sync
-uv pip install flash-attn --no-build-isolation
+pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja
+pip3 install --no-build-isolation -e '.[flash-attn]'
 
 # Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy
 python scripts/cutcrossentropy_install.py | sh

examples/colab-notebooks/colab-axolotl-example.ipynb

@@ -12,10 +12,10 @@
     "\n",
     "Axolotl is the most performant LLM post-training framework available, delivering faster training with efficient, consistent and stable performance. Train your workload and ship your product 30% faster; saving you both time and money.\n",
     "\n",
-    "- \u2b50 us on [GitHub](https://github.com/axolotl-ai-cloud/axolotl)\n",
-    "- \ud83d\udcdc Read the [Docs](http://docs.axolotl.ai/)\n",
-    "- \ud83d\udcac Chat with us on [Discord](https://discord.gg/mnpEYgRUmD)\n",
-    "- \ud83d\udcf0 Get updates on [X/Twitter](https://x.com/axolotl_ai)\n"
+    "- ⭐ us on [GitHub](https://github.com/axolotl-ai-cloud/axolotl)\n",
+    "- 📜 Read the [Docs](http://docs.axolotl.ai/)\n",
+    "- 💬 Chat with us on [Discord](https://discord.gg/mnpEYgRUmD)\n",
+    "- 📰 Get updates on [X/Twitter](https://x.com/axolotl_ai)\n"
    ]
   },
   {
@@ -39,8 +39,8 @@
    "source": [
     "%%capture\n",
     "# This step can take ~5-10 minutes to install dependencies\n",
-    "!uv pip install --no-build-isolation axolotl>=0.9.1\n!uv pip install flash-attn --no-build-isolation\n",
-    "!uv pip install \"cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@147ea28\""
+    "!pip install --no-build-isolation axolotl[flash-attn]>=0.9.1\n",
+    "!pip install \"cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@8a1a0ec\""
    ]
   },
   {
@@ -1371,7 +1371,7 @@
        "version_minor": 0
       },
       "text/plain": [
-       "VBox(children=(HTML(value='<center> <img\\nsrc=https://huggingface.co/front/assets/huggingface_logo-noborder.sv\u2026"
+       "VBox(children=(HTML(value='<center> <img\\nsrc=https://huggingface.co/front/assets/huggingface_logo-noborder.sv…"
       ]
      },
      "metadata": {},
@@ -1729,9 +1729,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_12815f401eba44658caa7b2e490137a8",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_30e02aa2d0d241979369e598287f2639",
-      "value": "Drop\u2007Samples\u2007with\u2007Zero\u2007Trainable\u2007Tokens\u2007(num_proc=2):\u2007100%"
+      "value": "Drop Samples with Zero Trainable Tokens (num_proc=2): 100%"
      }
     },
     "083f9cda8d754c168beee10d2f8955a2": {
@@ -1774,9 +1774,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_b195f160ca20442fadd8b5aed0ee41af",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_ca65e32eb52f48c09a84b33cb18f22cd",
-      "value": "\u200711.4M/11.4M\u2007[00:00&lt;00:00,\u200721.8MB/s]"
+      "value": " 11.4M/11.4M [00:00&lt;00:00, 21.8MB/s]"
      }
     },
     "0a46ad75c198463d843fb35e813642cb": {
@@ -1917,7 +1917,7 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_b1bea589efa14258a9982071b87938bf",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_590eef89881545aa8bbef9a8bbe7fb00",
       "value": "\n<b>Pro Tip:</b> If you don't already have one, you can create a dedicated\n'notebooks' token with 'write' access, that you can then easily reuse for all\nnotebooks. </center>"
      }
@@ -1938,9 +1938,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_bfcdbba993b74972a9e3e575f86908ff",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_6ebb2ec171414e47a14765505f64bb3c",
-      "value": "\u20073.84G/3.84G\u2007[00:09&lt;00:00,\u2007664MB/s]"
+      "value": " 3.84G/3.84G [00:09&lt;00:00, 664MB/s]"
      }
     },
     "0e936d9dbf9c4fdd86bbfe9730dedc47": {
@@ -2296,9 +2296,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_349eee9f56d64f0cba6fc24ff2c50c9b",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_7e5d3774060e4589aa65982da5ea4ef4",
-      "value": "\u20079985/9985\u2007[00:04&lt;00:00,\u20072604.11\u2007examples/s]"
+      "value": " 9985/9985 [00:04&lt;00:00, 2604.11 examples/s]"
      }
     },
     "16d1283741404b7bb319094c992fce01": {
@@ -2317,9 +2317,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_a4e5789584564049b83df7c6c54a3e08",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_ff3a94b146a948b6907f5d80c7157f99",
-      "value": "\u20079985/0\u2007[00:00&lt;00:00,\u200750763.46\u2007examples/s]"
+      "value": " 9985/0 [00:00&lt;00:00, 50763.46 examples/s]"
      }
     },
     "1811cda0644e4190a9469d1774435d82": {
@@ -2390,9 +2390,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_e366ae3fceec4566b9ed303d6c5f90af",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_5dd7d150dbe04f08b165ce7f2c27cd11",
-      "value": "model-00008-of-00008.safetensors:\u2007100%"
+      "value": "model-00008-of-00008.safetensors: 100%"
      }
     },
     "19127c7bb1554ccbac877059f9a82db0": {
@@ -2561,9 +2561,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_0dea5caa27384f5689e3cab51f558727",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_a6f48410b9964fefba0c3009a77dc838",
-      "value": "\u20079.68k/9.68k\u2007[00:00&lt;00:00,\u2007812kB/s]"
+      "value": " 9.68k/9.68k [00:00&lt;00:00, 812kB/s]"
      }
     },
     "1f7d30f71bbd4547a9150d21da071055": {
@@ -2634,9 +2634,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_f4a1795dc7514a718f478245f521f0ba",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_5e746eb25bbe416fb585fa24e79f5177",
-      "value": "model-00002-of-00008.safetensors:\u2007100%"
+      "value": "model-00002-of-00008.safetensors: 100%"
      }
     },
     "20352e5f58d24bb8b1f3940efd14fe4a": {
@@ -2707,9 +2707,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_1c6f1f10667545aaab958016ba7e2c94",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_e6e969610738449887259063967f82b0",
-      "value": "\u20072.78M/2.78M\u2007[00:00&lt;00:00,\u200717.8MB/s]"
+      "value": " 2.78M/2.78M [00:00&lt;00:00, 17.8MB/s]"
      }
     },
     "258b7c635c1045329d4669e48c46ccd5": {
@@ -3056,9 +3056,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_be724f04b03942b2a033a7e8898bb4fd",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_fcbab4d8dced41a18dfccce81e3a45a0",
-      "value": "model-00005-of-00008.safetensors:\u2007100%"
+      "value": "model-00005-of-00008.safetensors: 100%"
      }
     },
     "3036608c71904ce9ae4bb2a9fa8802d9": {
@@ -3077,9 +3077,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_5ca6be24acb548cea130bd58e9954c7c",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_5cfb02ee044b4011a378efa8b54a370f",
-      "value": "\u20073.96G/3.96G\u2007[00:10&lt;00:00,\u2007531MB/s]"
+      "value": " 3.96G/3.96G [00:10&lt;00:00, 531MB/s]"
      }
     },
     "30a81da86f8043eca301e86a8651201a": {
@@ -3629,9 +3629,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_8f5bd719974e41c3a8dd9a5b0d3d71e6",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_b87c84de30e84b3abf4871461fb9cbd3",
-      "value": "Loading\u2007checkpoint\u2007shards:\u2007100%"
+      "value": "Loading checkpoint shards: 100%"
      }
     },
     "41f3b32c2f6b4034ae7a3b9124e28bc7": {
@@ -3791,7 +3791,7 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_39789237703c4a418134243055c9cbf5",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_a3a945817f684328b34651fe052393ec",
       "value": "Connecting..."
      }
@@ -4077,9 +4077,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_4d468f96ec924681ad65eb671674b93e",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_ad7599de524549c48bf2d3124ad4b299",
-      "value": "Dropping\u2007Long\u2007Sequences\u2007(num_proc=2):\u2007100%"
+      "value": "Dropping Long Sequences (num_proc=2): 100%"
      }
     },
     "5ca240f31e6b44e3882c5eb37cd5a309": {
@@ -4471,9 +4471,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_5e18768f7ad6434ba8b8b8a2e853e204",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_bb33aec33a6447078c31bfd728942994",
-      "value": "\u2007728/728\u2007[00:00&lt;00:00,\u200720.3kB/s]"
+      "value": " 728/728 [00:00&lt;00:00, 20.3kB/s]"
      }
     },
     "62e302ebdad64aada0ffe64ae1c873f3": {
@@ -4636,9 +4636,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_81c3db71ac704280ad030072655f1537",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_042e091f75694c47aee761e760e76773",
-      "value": "\u20079985/9985\u2007[00:02&lt;00:00,\u20073977.47\u2007examples/s]"
+      "value": " 9985/9985 [00:02&lt;00:00, 3977.47 examples/s]"
      }
     },
     "67da6c4260574869aa24c3cbc1bc1654": {
@@ -4778,7 +4778,7 @@
       "description_tooltip": null,
       "disabled": false,
       "layout": "IPY_MODEL_2e257c8be2da40b4bb67a9e4ab6811f3",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_56e3768bef5a4b9db4168c5c17f509c2",
       "value": ""
      }
@@ -4823,9 +4823,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_41f3b32c2f6b4034ae7a3b9124e28bc7",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_a10d0a76010f4e508c65a9b69ebc5156",
-      "value": "Tokenizing\u2007Prompts\u2007(num_proc=2):\u2007100%"
+      "value": "Tokenizing Prompts (num_proc=2): 100%"
      }
     },
     "704f2f5a9b1c49d5a75a0025a5dda11b": {
@@ -5071,9 +5071,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_93a44a11aa4846fa8efc6c1413ef1627",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_a55060adc3564407ac81ad7297d34aaa",
-      "value": "train.jsonl:\u2007100%"
+      "value": "train.jsonl: 100%"
      }
     },
     "7be6f04c284e4326bb4ff3d301e7b3c6": {
@@ -5138,9 +5138,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_7fd44cf9ca6e4726bfd7ac21846d6a14",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_366a343b62fa47d8985a3bd464d99f9e",
-      "value": "config.json:\u2007100%"
+      "value": "config.json: 100%"
      }
     },
     "7cd0b85ebd204b7aba908417811ce4e0": {
@@ -5339,9 +5339,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_67da6c4260574869aa24c3cbc1bc1654",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_94b9088614464f60a203de39dbcae853",
-      "value": "\u20078/8\u2007[01:47&lt;00:00,\u200711.64s/it]"
+      "value": " 8/8 [01:47&lt;00:00, 11.64s/it]"
      }
     },
     "823f1c78f15043e38bbd4dca3932a86a": {
@@ -5488,7 +5488,7 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_8640ac440fbc4644b9a3af7ba3ae7183",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_5cea7996f02040b187ece0bb2d6a8d1f",
       "value": "<center> <img\nsrc=https://huggingface.co/front/assets/huggingface_logo-noborder.svg\nalt='Hugging Face'> <br> Copy a token from <a\nhref=\"https://huggingface.co/settings/tokens\" target=\"_blank\">your Hugging Face\ntokens page</a> and paste it below. <br> Immediately click login after copying\nyour token or it might be stored in plain text in this notebook file. </center>"
      }
@@ -5509,9 +5509,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_ef223e8504b64e3592589880326aaf41",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_598da69727bd4fb8b1caf465ac736d7a",
-      "value": "\u20071.67M/1.67M\u2007[00:00&lt;00:00,\u200719.0MB/s]"
+      "value": " 1.67M/1.67M [00:00&lt;00:00, 19.0MB/s]"
      }
     },
     "897b77a56c09479bb11d7f2a30997e55": {
@@ -5717,9 +5717,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_37de928300e34184881039378bd75e7f",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_0e936d9dbf9c4fdd86bbfe9730dedc47",
-      "value": "\u20073.96G/3.96G\u2007[00:13&lt;00:00,\u2007273MB/s]"
+      "value": " 3.96G/3.96G [00:13&lt;00:00, 273MB/s]"
      }
     },
     "936d04b5fe1b4c63bf0b080e423d051b": {
@@ -6050,9 +6050,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_d955dcaa0e944e719f3a06139dd54a03",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_d3de2662c7964f1ba96e58da382af720",
-      "value": "merges.txt:\u2007100%"
+      "value": "merges.txt: 100%"
      }
     },
     "9cd5211b5d8b457aa0002f1d17b80028": {
@@ -6071,9 +6071,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_6932489232ec4ab18a160b1e7fbcdfe1",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_4540927d98f54466b434ba4c0edf045d",
-      "value": "model-00007-of-00008.safetensors:\u2007100%"
+      "value": "model-00007-of-00008.safetensors: 100%"
      }
     },
     "9d4897eefb5f48259ffb2d23e332f752": {
@@ -6303,9 +6303,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_3aaecbf540f54a2db9ab0931e3b1fe57",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_9e333ed3b5014069ac1dd969255dd591",
-      "value": "\u2007239/239\u2007[00:00&lt;00:00,\u200730.9kB/s]"
+      "value": " 239/239 [00:00&lt;00:00, 30.9kB/s]"
      }
     },
     "a20927bf5f2c41f58c1e31ac858ab36c": {
@@ -6324,9 +6324,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_1811cda0644e4190a9469d1774435d82",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_35c811d2ae8e43f3b5cecbdd3cfa857f",
-      "value": "tokenizer.json:\u2007100%"
+      "value": "tokenizer.json: 100%"
      }
     },
     "a3a945817f684328b34651fe052393ec": {
@@ -6360,9 +6360,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_ed5ca967ad5342929e578ac6aa4dc4c0",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_af401d117d5047629d3a6e2361757b62",
-      "value": "model-00001-of-00008.safetensors:\u2007100%"
+      "value": "model-00001-of-00008.safetensors: 100%"
      }
     },
     "a4e5789584564049b83df7c6c54a3e08": {
@@ -6494,9 +6494,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_fa1282ccc7544e4f818e2f03ccffe4a5",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_bbbf575d2a4b4c6ea8389be79b2a6039",
-      "value": "model.safetensors.index.json:\u2007100%"
+      "value": "model.safetensors.index.json: 100%"
      }
     },
     "ab93eabd7cea4b94b4b7a387f101e8a1": {
@@ -6582,9 +6582,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_62e302ebdad64aada0ffe64ae1c873f3",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_bd1b0dfed6d34d16af33a4a58330f5ec",
-      "value": "Saving\u2007the\u2007dataset\u2007(1/1\u2007shards):\u2007100%"
+      "value": "Saving the dataset (1/1 shards): 100%"
      }
     },
     "ad7599de524549c48bf2d3124ad4b299": {
@@ -6967,9 +6967,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_2b3a2659b12244bd8548320320016dbf",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_0cd7efffbb3c4c4b972e63749f61ab97",
-      "value": "Generating\u2007train\u2007split:\u2007"
+      "value": "Generating train split: "
      }
     },
     "b87c84de30e84b3abf4871461fb9cbd3": {
@@ -7085,9 +7085,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_0f480e3a0b0a45d2a2d2dec3cad923f3",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_fcb30372e7404c5d8a1ad4df91e6c7b2",
-      "value": "\u20071.91G/1.91G\u2007[00:05&lt;00:00,\u2007444MB/s]"
+      "value": " 1.91G/1.91G [00:05&lt;00:00, 444MB/s]"
      }
     },
     "bd1b0dfed6d34d16af33a4a58330f5ec": {
@@ -7325,9 +7325,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_158c8b85dbf34de6a94b4e35e2fc7d5a",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_0b4c9753a7cb4354b8e5f187e6e1ad7c",
-      "value": "\u20073.96G/3.96G\u2007[00:15&lt;00:00,\u2007564MB/s]"
+      "value": " 3.96G/3.96G [00:15&lt;00:00, 564MB/s]"
      }
     },
     "c0991cf63ee6458b96e9a75e7a88b61a": {
@@ -7346,9 +7346,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_ed28e2e0410d4e0b855467e798e53d66",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_d93f134f802b4b69b575bdaf07dbd27c",
-      "value": "tokenizer_config.json:\u2007100%"
+      "value": "tokenizer_config.json: 100%"
      }
     },
     "c12ea43372ac4d57bb9605f1a429b397": {
@@ -7581,9 +7581,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_8bc9d8ba866c442b9118d9630009939c",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_9f56a2d9979c4bd8928c644c22c3ecdf",
-      "value": "model-00003-of-00008.safetensors:\u2007100%"
+      "value": "model-00003-of-00008.safetensors: 100%"
      }
     },
     "c6164e05a1914ae48083db9ad7f4ef7c": {
@@ -7694,9 +7694,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_e40d1c1ac9494b3bade9858324e7ffdf",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_d65b6b060d9845779299491ac5599c31",
-      "value": "\u20079985/9985\u2007[01:04&lt;00:00,\u2007189.08\u2007examples/s]"
+      "value": " 9985/9985 [01:04&lt;00:00, 189.08 examples/s]"
      }
     },
     "c7433acd3c4841e6958ae8f7e87b1808": {
@@ -7737,9 +7737,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_0077aedc3d174560bce924ee89e9c006",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_00321cce58884f6f9b3855a21fcd9187",
-      "value": "Add\u2007position_id\u2007column\u2007(Sample\u2007Packing)\u2007(num_proc=2):\u2007100%"
+      "value": "Add position_id column (Sample Packing) (num_proc=2): 100%"
      }
     },
     "ca65e32eb52f48c09a84b33cb18f22cd": {
@@ -8162,9 +8162,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_63580b6fb30642479fe3000915bf551a",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_8f726dbfb45d4528afa33e36a6313267",
-      "value": "\u200727.3M/27.3M\u2007[00:00&lt;00:00,\u200731.0MB/s]"
+      "value": " 27.3M/27.3M [00:00&lt;00:00, 31.0MB/s]"
      }
     },
     "d43c6df07ddb466587807d6dbe1ff614": {
@@ -8183,9 +8183,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_8c4d4fc5a30f4e7cb3be53fe2adda33d",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_e90658f4bcb642baa78426012f863152",
-      "value": "model-00004-of-00008.safetensors:\u2007100%"
+      "value": "model-00004-of-00008.safetensors: 100%"
      }
     },
     "d65b6b060d9845779299491ac5599c31": {
@@ -8474,9 +8474,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_34cf3df51fbc41cabfdbba153c007f0e",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_ac764024cf1c4e08ba7749afd2cd20ac",
-      "value": "vocab.json:\u2007100%"
+      "value": "vocab.json: 100%"
      }
     },
     "dfd2a2649b8341ef913207526708aff1": {
@@ -8669,9 +8669,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_c6164e05a1914ae48083db9ad7f4ef7c",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_813621384dc748b0ad06775e22761c0b",
-      "value": "\u20079985/9985\u2007[00:03&lt;00:00,\u20073622.89\u2007examples/s]"
+      "value": " 9985/9985 [00:03&lt;00:00, 3622.89 examples/s]"
      }
     },
     "e400cbf14bcc446a9d33b210cd93550b": {
@@ -9065,9 +9065,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_fba7aa824b38467ab3061b226114cdec",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_f3075dccbd2747b4a7913b66f44f2596",
-      "value": "\u20073.96G/3.96G\u2007[00:13&lt;00:00,\u2007398MB/s]"
+      "value": " 3.96G/3.96G [00:13&lt;00:00, 398MB/s]"
      }
     },
     "ec030fc3c346426f9abc3a89892258d3": {
@@ -9110,9 +9110,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_936d04b5fe1b4c63bf0b080e423d051b",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_f1cef8e8dc2646fb9fd09f3b09081074",
-      "value": "\u200736.5k/36.5k\u2007[00:00&lt;00:00,\u20074.32MB/s]"
+      "value": " 36.5k/36.5k [00:00&lt;00:00, 4.32MB/s]"
      }
     },
     "ed28e2e0410d4e0b855467e798e53d66": {
@@ -9422,9 +9422,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_735d4f225b24414294fc1b213c61223c",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_5e5e15b0569b474c9620083b3ec6af55",
-      "value": "generation_config.json:\u2007100%"
+      "value": "generation_config.json: 100%"
      }
     },
     "f4667818b9d34a09891cd727a429a610": {
@@ -9443,9 +9443,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_4b27c267393640f28f6eae0875bd2ed9",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_9858cb74a09748a39e8149baac96702c",
-      "value": "\u20073.96G/3.96G\u2007[00:11&lt;00:00,\u2007457MB/s]"
+      "value": " 3.96G/3.96G [00:11&lt;00:00, 457MB/s]"
      }
     },
     "f4a1795dc7514a718f478245f521f0ba": {
@@ -9830,9 +9830,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_d1f9b10c130542f094c8fd3d1e23b5e9",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_e575d87a7efe4ec7b1efde489839d4a6",
-      "value": "model-00006-of-00008.safetensors:\u2007100%"
+      "value": "model-00006-of-00008.safetensors: 100%"
      }
     },
     "fe18bba7f3fb4c31bf840541f36b3425": {
@@ -9873,9 +9873,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_e5a82df528bb4e408797a3b6c2758f4a",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_f113ebd8c1c34806bea4dd7ed3035173",
-      "value": "\u20079985/9985\u2007[00:00&lt;00:00,\u200744264.88\u2007examples/s]"
+      "value": " 9985/9985 [00:00&lt;00:00, 44264.88 examples/s]"
      }
     },
     "fea1b70fb46745feb5111b3929175b5d": {
@@ -9931,9 +9931,9 @@
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_ab93eabd7cea4b94b4b7a387f101e8a1",
-      "placeholder": "\u200b",
+      "placeholder": "​",
       "style": "IPY_MODEL_704f2f5a9b1c49d5a75a0025a5dda11b",
-      "value": "\u20073.96G/3.96G\u2007[00:12&lt;00:00,\u2007656MB/s]"
+      "value": " 3.96G/3.96G [00:12&lt;00:00, 656MB/s]"
      }
     }
    }

examples/devstral/README.md

@@ -16,13 +16,8 @@ Thanks to the team at MistralAI for giving us early access to prepare for this r
 
 ```bash
 # Ensure you have Pytorch installed (Pytorch 2.6.0 min)
-# Option A: manage dependencies in your project
-uv add 'axolotl>=0.12.0'
-uv pip install flash-attn --no-build-isolation
-
-# Option B: quick install
-uv pip install 'axolotl>=0.12.0'
-uv pip install flash-attn --no-build-isolation
+pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja
+pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
 ```
 
 2. Install [Cut Cross Entropy](https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy) to reduce training VRAM usage

examples/gemma3/gemma-3-1b-qlora.yml

@@ -1,7 +1,7 @@
 base_model: google/gemma-3-1b-it
-# optionally might have model_type or tokenizer_type
-model_type: AutoModelForCausalLM
-tokenizer_type: AutoTokenizer
+
+model_type: Gemma3ForCausalLM
+
 # Automatically upload checkpoint and final model to HF
 # hub_model_id: username/custom_model_name
 

examples/gemma3/gemma-3-270m-qlora.yml

@@ -1,7 +1,7 @@
 base_model: google/gemma-3-270m-it
-# optionally might have model_type or tokenizer_type
-model_type: AutoModelForCausalLM
-tokenizer_type: AutoTokenizer
+
+model_type: Gemma3ForCausalLM
+
 # Automatically upload checkpoint and final model to HF
 # hub_model_id: username/custom_model_name
 

examples/gemma3/gemma-3-4b-qlora.yml

@@ -1,5 +1,8 @@
 base_model: google/gemma-3-4b-it
 
+# Need to set else transformers tries to load vision too
+model_type: Gemma3ForCausalLM
+
 load_in_4bit: true
 
 # gemma3 doesn't seem to play nice with ddp

examples/gemma3n/README.md

@@ -10,22 +10,17 @@ Gemma-3n is a family of multimodal models from Google found on [HuggingFace](htt
 
 ```bash
 # Ensure you have Pytorch installed (Pytorch 2.6.0 min)
-# Option A: manage dependencies in your project
-uv add 'axolotl>=0.12.0'
-uv pip install flash-attn --no-build-isolation
-
-# Option B: quick install
-uv pip install 'axolotl>=0.12.0'
-uv pip install flash-attn --no-build-isolation
+pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja
+pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
 ```
 
 2. In addition to Axolotl's requirements, Gemma-3n requires:
 
 ```bash
-uv pip install timm==1.0.17
+pip3 install timm==1.0.17
 
 # for loading audio data
-uv pip install librosa==0.11.0
+pip3 install librosa==0.11.0
 ```
 
 3. Download sample dataset files

examples/gpt-oss/README.md

@@ -2,6 +2,8 @@
 
 [GPT-OSS](https://huggingface.co/collections/openai/gpt-oss-68911959590a1634ba11c7a4) are a family of open-weight MoE models trained by OpenAI, released in August 2025. There are two variants: 20B and 120B.
 
+In October 2025, OpenAI released safeguard models built upon GPT-OSS called [GPT-OSS-Safeguard](https://huggingface.co/collections/openai/gpt-oss-safeguard). They use the same architecture, so the same examples below can be re-used.
+
 This guide shows how to fine-tune it with Axolotl with multi-turn conversations and proper masking.
 
 ## Getting started
@@ -12,13 +14,8 @@ This guide shows how to fine-tune it with Axolotl with multi-turn conversations
 
 ```bash
 # Ensure you have Pytorch installed (Pytorch 2.6.0 min)
-# Option A: manage dependencies in your project
-uv add 'axolotl>=0.12.0'
-uv pip install flash-attn --no-build-isolation
-
-# Option B: quick install
-uv pip install 'axolotl>=0.12.0'
-uv pip install flash-attn --no-build-isolation
+pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja
+pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
 ```
 
 2. Choose one of the following configs below for training the 20B model. (for 120B, see [below](#training-120b))
@@ -69,6 +66,16 @@ axolotl merge-sharded-fsdp-weights examples/gpt-oss/gpt-oss-120b-fft-fsdp2-offlo
 mv ./outputs/gpt-oss-out/merged/* ./outputs/gpt-oss-out/
 ```
 
+### How to set reasoning_effort in template?
+
+The harmony template has a feature to set the `reasoning_effort` during prompt building. The default is `medium`. If you would like to adjust this, you can add the following to your config:
+
+```yaml
+chat_template_kwargs:
+  reasoning_effort: "high"  # low | medium | high
+```
+
+Currently, this applies globally. There is no method to apply per sample yet. If you are interested in adding this, please feel free to create an Issue to discuss.
 
 ### Inferencing your fine-tuned model
 
@@ -80,7 +87,7 @@ for more information about using a special vllm-openai docker image for inferenc
 Optionally, vLLM can be installed from nightly:
 
 ```bash
-uv pip install --no-build-isolation --pre -U vllm --extra-index-url https://wheels.vllm.ai/nightly
+pip install --no-build-isolation --pre -U vllm --extra-index-url https://wheels.vllm.ai/nightly
 ```
 and the vLLM server can be started with the following command (modify `--tensor-parallel-size 8` to match your environment):
 ```bash

examples/gpt-oss/gpt-oss-safeguard-20b-sft-lora-singlegpu.yaml

@@ -0,0 +1,67 @@
+base_model: openai/gpt-oss-safeguard-20b
+use_kernels: true
+model_quantization_config: Mxfp4Config
+model_quantization_config_kwargs:
+  dequantize: true
+
+plugins:
+  - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
+
+experimental_skip_move_to_device: true  # prevent OOM by not putting model to GPU before sharding
+
+datasets:
+  - path: HuggingFaceH4/Multilingual-Thinking
+    type: chat_template
+    field_thinking: thinking
+    template_thinking_key: thinking
+
+dataset_prepared_path: last_run_prepared
+val_set_size: 0
+output_dir: ./outputs/gpt-oss-safeguard-out/
+
+sequence_len: 4096
+sample_packing: true
+
+adapter: lora
+lora_r: 8
+lora_alpha: 16
+lora_dropout: 0.0  # dropout not supported when using LoRA over expert parameters
+lora_target_linear: true
+
+# TODO: not supported for now, see peft#2710
+#lora_target_parameters:  # target the experts in the last two layers
+#  - "22._checkpoint_wrapped_module.mlp.experts.gate_up_proj"
+#  - "22._checkpoint_wrapped_module.mlp.experts.down_proj"
+#  - "23._checkpoint_wrapped_module.mlp.experts.gate_up_proj"
+#  - "23._checkpoint_wrapped_module.mlp.experts.down_proj"
+
+wandb_project:
+wandb_entity:
+wandb_watch:
+wandb_name:
+wandb_log_model:
+
+gradient_accumulation_steps: 8
+micro_batch_size: 1
+num_epochs: 1
+
+optimizer: adamw_torch_8bit
+lr_scheduler: constant_with_warmup
+learning_rate: 2e-4
+
+bf16: true
+tf32: true
+
+flash_attention: true
+attn_implementation: kernels-community/vllm-flash-attn3  # this is not needed if using flash_attn >= 2.8.3
+
+gradient_checkpointing: true
+activation_offloading: true
+
+logging_steps: 1
+saves_per_epoch: 1
+warmup_ratio: 0.1
+
+special_tokens:
+eot_tokens:
+  - "<|end|>"

examples/hunyuan/README.md

@@ -13,8 +13,8 @@ Tencent released a family of opensource models called HunYuan with varying param
 git clone https://github.com/axolotl-ai-cloud/axolotl.git
 cd axolotl
 
-uv sync
-uv pip install flash-attn --no-build-isolation
+pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja
+pip3 install --no-build-isolation -e '.[flash-attn]'
 
 # Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy
 python scripts/cutcrossentropy_install.py | sh

examples/llama-3/3b-fp8-fsdp2.yaml

@@ -29,7 +29,7 @@ flex_attention: true
 flex_attn_compile_kwargs:
   dynamic: false
   mode: max-autotune-no-cudagraphs
-
+save_strategy: no
 torch_compile: true
 
 wandb_project:

examples/llama-3/opentelemetry-qlora.yml

@@ -0,0 +1,50 @@
+base_model: NousResearch/Llama-3.2-1B
+model_type: AutoModelForCausalLM
+tokenizer_type: AutoTokenizer
+
+load_in_4bit: true
+
+datasets:
+  - path: mhenrichsen/alpaca_2k_test
+    type: alpaca
+
+output_dir: ./outputs/opentelemetry-example
+
+adapter: qlora
+sequence_len: 512
+sample_packing: false
+
+lora_r: 32
+lora_alpha: 16
+lora_dropout: 0.05
+lora_target_linear: true
+
+# OpenTelemetry Configuration
+use_otel_metrics: true
+otel_metrics_host: "localhost"
+otel_metrics_port: 8000
+
+# Disable WandB
+use_wandb: false
+
+gradient_accumulation_steps: 4
+micro_batch_size: 2
+num_epochs: 1
+optimizer: paged_adamw_32bit
+lr_scheduler: cosine
+learning_rate: 0.0002
+
+bf16: auto
+tf32: false
+
+gradient_checkpointing: true
+logging_steps: 1
+flash_attention: false
+
+warmup_ratio: 0.1
+evals_per_epoch: 2
+saves_per_epoch: 1
+weight_decay: 0.0
+
+special_tokens:
+  pad_token: "<|end_of_text|>"

examples/magistral/README.md

@@ -13,14 +13,9 @@ Thanks to the team at MistralAI for giving us early access to prepare for these
     Here is an example of how to install from pip:
 
 ```bash
-# Ensure you have PyTorch installed (PyTorch 2.6.0 min)
-# Option A: manage dependencies in your project
-uv add 'axolotl>=0.12.0'
-uv pip install flash-attn --no-build-isolation
-
-# Option B: quick install
-uv pip install 'axolotl>=0.12.0'
-uv pip install flash-attn --no-build-isolation
+# Ensure you have Pytorch installed (Pytorch 2.6.0 min)
+pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja
+pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
 ```
 
 2. Install [Cut Cross Entropy](https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy) to reduce training VRAM usage

examples/magistral/think/README.md

@@ -12,7 +12,7 @@ Before starting, ensure you have:
 Run the thinking model fine-tuning:
 
 ```bash
-axolotl train magistral-small-think-qlora.yaml
+axolotl train examples/magistral/think/magistral-small-think-qlora.yaml
 ```
 
 This config uses about 19.1 GiB VRAM.

examples/magistral/vision/README.md

@@ -21,7 +21,7 @@ Before starting, ensure you have:
 
 3. Run the fine-tuning:
    ```bash
-   axolotl train magistral-small-vision-24B-qlora.yml
+   axolotl train examples/magistral/vision/magistral-small-vision-24B-qlora.yml
    ```
 
 This config uses about 17GiB VRAM.

examples/mistral/mistral-small/README.md

@@ -0,0 +1,51 @@
+# Mistral Small 3.1/3.2 Fine-tuning
+
+This guide covers fine-tuning [Mistral Small 3.1](mistralai/Mistral-Small-3.1-24B-Instruct-2503) and [Mistral Small 3.2](mistralai/Mistral-Small-3.2-24B-Instruct-2506) with vision capabilities using Axolotl.
+
+## Prerequisites
+
+Before starting, ensure you have:
+- Installed Axolotl (see [Installation docs](https://docs.axolotl.ai/docs/installation.html))
+
+## Getting Started
+
+1. Install the required vision lib:
+    ```bash
+    pip install 'mistral-common[opencv]==1.8.5'
+    ```
+
+2. Download the example dataset image:
+   ```bash
+   wget https://huggingface.co/datasets/Nanobit/text-vision-2k-test/resolve/main/African_elephant.jpg
+   ```
+
+3. Run the fine-tuning:
+   ```bash
+   axolotl train examples/mistral/mistral-small/mistral-small-3.1-24B-lora.yml
+   ```
+
+This config uses about 29.4 GiB VRAM.
+
+## Dataset Format
+
+The vision model requires multi-modal dataset format as documented [here](https://docs.axolotl.ai/docs/multimodal.html#dataset-format).
+
+One exception is that, passing `"image": PIL.Image` is not supported. MistralTokenizer only supports `path`, `url`, and `base64` for now.
+
+Example:
+```json
+{
+    "messages": [
+        {"role": "system", "content": [{ "type": "text", "text": "{SYSTEM_PROMPT}"}]},
+        {"role": "user", "content": [
+            { "type": "text", "text": "What's in this image?"},
+            {"type": "image", "path": "path/to/image.jpg" }
+        ]},
+        {"role": "assistant", "content": [{ "type": "text", "text": "..." }]},
+    ],
+}
+```
+
+## Limitations
+
+- Sample Packing is not supported for multi-modality training currently.

examples/mistral/mistral-small/mistral-small-3.1-24B-lora.yml

@@ -39,7 +39,7 @@ wandb_name:
 wandb_log_model:
 
 gradient_accumulation_steps: 1
-micro_batch_size: 1
+micro_batch_size: 2
 num_epochs: 1
 optimizer: adamw_bnb_8bit
 lr_scheduler: cosine

examples/qwen3-next/README.md

@@ -15,8 +15,8 @@ This guide shows how to fine-tune it with Axolotl with multi-turn conversations
 git clone https://github.com/axolotl-ai-cloud/axolotl.git
 cd axolotl
 
-uv sync
-uv pip install flash-attn --no-build-isolation
+pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja
+pip3 install --no-build-isolation -e '.[flash-attn]'
 
 # Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy
 python scripts/cutcrossentropy_install.py | sh
@@ -24,12 +24,12 @@ python scripts/cutcrossentropy_install.py | sh
 
 2. Install Qwen3-Next transformers commit
 ```bash
-uv pip uninstall -y transformers && uv pip install "git+https://github.com/huggingface/transformers.git@b9282355bea846b54ed850a066901496b19da654"
+pip3 uninstall -y transformers && pip3 install "git+https://github.com/huggingface/transformers.git@b9282355bea846b54ed850a066901496b19da654"
 ```
 
 3. Install FLA for improved performance
 ```bash
-uv pip uninstall -y causal-conv1d && uv pip install flash-linear-attention==0.3.2
+pip3 uninstall -y causal-conv1d && pip3 install flash-linear-attention==0.3.2
 ```
 
 4. Run the finetuning example:

examples/seed-oss/README.md

@@ -15,8 +15,8 @@ This guide shows how to fine-tune it with Axolotl with multi-turn conversations
 git clone https://github.com/axolotl-ai-cloud/axolotl.git
 cd axolotl
 
-uv sync --extra deepspeed
-uv pip install flash-attn --no-build-isolation
+pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja
+pip3 install --no-build-isolation -e '.[flash-attn]'
 
 # Install Cut Cross Entropy
 python scripts/cutcrossentropy_install.py | sh

examples/smolvlm2/README.md

@@ -13,19 +13,14 @@ This guide shows how to fine-tune SmolVLM2 models with Axolotl.
     Here is an example of how to install from pip:
     ```bash
     # Ensure you have a compatible version of Pytorch installed
-    # Option A: manage dependencies in your project
-    uv add 'axolotl>=0.12.0'
-    uv pip install flash-attn --no-build-isolation
-
-    # Option B: quick install
-    uv pip install 'axolotl>=0.12.0'
-    uv pip install flash-attn --no-build-isolation
+    pip3 install packaging setuptools wheel ninja
+    pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
     ```
 
 2. Install an extra dependency:
 
     ```bash
-    uv pip install num2words==0.5.14
+    pip3 install num2words==0.5.14
     ```
 
 3.  Run the finetuning example:

examples/voxtral/README.md

@@ -12,21 +12,16 @@ Thanks to the team at MistralAI for giving us early access to prepare for this r
 
 ```bash
 # Ensure you have Pytorch installed (Pytorch 2.6.0 min)
-# Option A: manage dependencies in your project
-uv add 'axolotl>=0.12.0'
-uv pip install flash-attn --no-build-isolation
-
-# Option B: quick install
-uv pip install 'axolotl>=0.12.0'
-uv pip install flash-attn --no-build-isolation
+pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja
+pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
 ```
 
 2. Please install the below.
 
 ```bash
 # audio
-uv pip install librosa==0.11.0
-uv pip install 'mistral_common[audio]==1.8.3'
+pip3 install librosa==0.11.0
+pip3 install 'mistral_common[audio]==1.8.3'
 
 # Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy
 python scripts/cutcrossentropy_install.py | sh

pyproject.toml

@@ -1,131 +1,14 @@
 [build-system]
-requires = ["setuptools>=64", "wheel", "setuptools_scm>=8"]
+requires = ["setuptools>=64", "wheel", "setuptools_scm>=8", "packaging==23.2"]
 build-backend = "setuptools.build_meta"
 
 [project]
 name = "axolotl"
-dynamic = ["version"]
+dynamic = ["version", "dependencies", "optional-dependencies"]
 description = "LLM Trainer"
 readme = "README.md"
-requires-python = ">=3.10,<3.13"
-license = {text = "Apache-2.0"}
-authors = [
-    {name = "Axolotl AI"},
-]
-maintainers = [
-    {name = "Axolotl AI"},
-]
-classifiers = [
-    "Development Status :: 4 - Beta",
-    "License :: OSI Approved :: Apache Software License",
-    "Programming Language :: Python :: 3",
-    "Programming Language :: Python :: 3.10",
-    "Programming Language :: Python :: 3.11",
-    "Programming Language :: Python :: 3.12",
-]
-
-dependencies = [
-    "torch>=2.6.0",
-    "packaging>=23.2",
-    "huggingface_hub>=0.33.0",
-    "peft==0.17.0",
-    "transformers==4.56.1",
-    "tokenizers>=0.21.1",
-    "accelerate==1.10.1",
-    "datasets==4.0.0",
-    "trl==0.23.0",
-    "hf_xet==1.1.5",
-    "kernels==0.9.0",
-    "trackio",
-    "optimum==1.16.2",
-    "hf_transfer",
-    "sentencepiece",
-    "gradio==5.41.1",
-    "modal==1.0.2",
-    "pydantic>=2.10.6",
-    "addict",
-    "fire",
-    "PyYAML>=6.0",
-    "requests",
-    "wandb",
-    "einops",
-    "colorama",
-    "numba",
-    "numpy>=1.24.4,<3.0",
-    "evaluate==0.4.1",
-    "scipy",
-    "scikit-learn>=1.7.0",
-    "nvidia-ml-py==12.560.30",
-    "art",
-    "tensorboard",
-    "python-dotenv==1.0.1",
-    "s3fs>=2024.5.0",
-    "gcsfs>=2024.5.0",
-    "adlfs>=2024.5.0",
-    "ocifs==1.3.2",
-    "zstandard>=0.23.0",
-    "fastcore",
-    "lm_eval==0.4.7",
-    "langdetect==1.0.9",
-    "immutabledict==4.2.0",
-    "antlr4-python3-runtime==4.13.2",
-    "schedulefree==1.4.1",
-    "mistral-common==1.8.5",
-
-    # Axolotl contribs
-    "axolotl-contribs-lgpl @ git+https://github.com/axolotl-ai-cloud/axolotl-contribs-lgpl.git@numpy",
-    "axolotl-contribs-mit==0.0.5",
-
-    # Platform-specific dependencies (Linux by default, excluded on macOS)
-    "triton>=3.0.0 ; sys_platform != 'darwin'",
-    "xformers>=0.0.28 ; sys_platform != 'darwin'",
-    "autoawq==0.2.7.post3 ; sys_platform != 'darwin'",
-    "liger-kernel==0.6.1 ; sys_platform != 'darwin'",
-    "torchao==0.13.0 ; sys_platform != 'darwin'",
-    "bitsandbytes==0.47.0 ; sys_platform != 'darwin'",
-    "deepspeed>=0.17.5 ; sys_platform != 'darwin'",
-    "deepspeed-kernels ; sys_platform != 'darwin'",
-]
-
-[project.optional-dependencies]
-ring-flash-attn = [
-    "ring-flash-attn>=0.1.7",
-    "yunchang==0.6.0",
-]
-mamba-ssm = ["mamba-ssm>=2.2.0", "causal_conv1d>=1.4.0",]
-gptqmodel = ["gptqmodel>=4.0.0"]
-mlflow = ["mlflow"]
-galore = ["galore_torch"]
-apollo = ["apollo-torch"]
-optimizers = [
-    "galore_torch",
-    "apollo-torch",
-    "lomo-optim==0.1.1",
-    "torch-optimi==0.2.1",
-    "came_pytorch==0.1.3",
-]
-ray = ["ray[train]"]
-vllm = ["vllm>=0.10.0"]
-llmcompressor = ["llmcompressor>=0.5.1"]
-fbgemm-gpu = ["fbgemm-gpu-genai>=1.2.0"]
-dev = [
-    "pytest",
-    "pytest-cov",
-    "pytest-retry",
-    "pytest-sugar",
-    "pytest-xdist",
-    "codecov",
-    "codecov-cli",
-    "tbparse",
-    "ruff",
-    "mypy",
-    "pre-commit",
-    "types-requests",
-    "quartodoc",
-    "jupyter",
-    "blobfile",
-    "tiktoken",
-]
+requires-python = ">=3.10"
+# license = "Apache-2.0"
 
 [project.scripts]
 axolotl = "axolotl.cli.main:main"
@@ -134,20 +17,15 @@ axolotl = "axolotl.cli.main:main"
 Homepage = "https://axolotl.ai/"
 Documentation = "https://docs.axolotl.ai/"
 Repository = "https://github.com/axolotl-ai-cloud/axolotl.git"
-Issues = "https://github.com/axolotl-ai-cloud/axolotl/issues"
-
-[tool.setuptools]
-package-dir = {"" = "src"}
-include-package-data = true
-
-[tool.setuptools.packages.find]
-where = ["src"]
-
-[tool.setuptools.package-data]
-"*" = ["*.yaml", "*.yml", "*.json"]
 
 [tool.setuptools_scm]
-write_to = "src/axolotl/_version.py"
+
+[tool.setuptools]
+py-modules = ["setuptools_axolotl_dynamic_dependencies"]
+include-package-data = true
+
+[tool.setuptools.cmdclass]
+build_py = "setuptools_axolotl_dynamic_dependencies.BuildPyCommand"
 
 [tool.ruff]
 line-length = 88
@@ -179,60 +57,3 @@ indent-style = "space"
 skip-magic-trailing-comma = false
 line-ending = "auto"
 docstring-code-format = false
-
-[tool.mypy]
-python_version = "3.11"
-warn_return_any = true
-warn_unused_configs = true
-ignore_missing_imports = true
-
-[tool.pytest.ini_options]
-testpaths = ["tests"]
-python_files = ["test_*.py", "*_test.py"]
-addopts = "-v --tb=short"
-
-# UV specific configuration
-[tool.uv]
-prerelease = "allow"
-default-groups = ["default"]
-conflicts = [
-    [
-        { group = "default" },
-        { extra = "vllm" },
-    ],
-]
-
-[dependency-groups]
-default = ["torch>=2.6.0"]
-dev = [
-    "pytest",
-    "pytest-cov",
-    "pytest-retry",
-    "pytest-sugar",
-    "pytest-xdist",
-    "codecov",
-    "codecov-cli",
-    "tbparse",
-    "ruff",
-    "mypy",
-    "pre-commit",
-    "types-requests",
-    "quartodoc",
-    "jupyter",
-    "blobfile",
-    "tiktoken",
-]
-
-[[tool.uv.index]]
-name = "autogptq"
-url = "https://huggingface.github.io/autogptq-index/whl/"
-
-[tool.uv.extra-build-dependencies]
-mamba-ssm = ["torch", "causal_conv1d"]
-gptqmodel = [
-    { requirement = "torch", match-runtime = true },
-]
-autoawq = ["torch"]
-triton = ["torch"]
-bitsandbytes = ["torch"]
-grpclib = ["wheel"]

requirements-dev.txt

@@ -0,0 +1,8 @@
+black
+mypy
+pre-commit
+types-requests
+quartodoc
+jupyter
+blobfile
+tiktoken

requirements-tests.txt

@@ -0,0 +1,8 @@
+codecov
+codecov-cli
+pytest
+pytest-cov
+pytest-retry
+pytest-sugar
+pytest-xdist
+tbparse

requirements.txt

@@ -0,0 +1,72 @@
+--extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/
+
+# START section of dependencies that don't install on Darwin/MacOS
+bitsandbytes==0.47.0
+triton>=3.0.0
+mamba-ssm==1.2.0.post1
+xformers>=0.0.23.post1
+liger-kernel==0.6.3
+# END section
+
+packaging==23.2
+
+huggingface_hub>=0.36.0
+peft>=0.17.1
+tokenizers>=0.21.1
+transformers==4.57.1
+accelerate==1.10.1
+datasets==4.3.0
+deepspeed>=0.17.0
+trl==0.24.0
+hf_xet==1.2.0
+kernels>=0.9.0
+trackio
+
+optimum==1.16.2
+hf_transfer
+sentencepiece
+gradio==5.49.1
+
+modal==1.0.2
+pydantic>=2.10.6
+addict
+fire
+PyYAML>=6.0
+requests
+wandb
+einops
+colorama
+numba>=0.61.2
+numpy>=2.2.6
+
+# qlora things
+evaluate==0.4.1
+scipy
+scikit-learn==1.4.2
+nvidia-ml-py==12.560.30
+art
+tensorboard
+python-dotenv==1.0.1
+
+# remote filesystems
+s3fs>=2024.5.0
+gcsfs>=2025.3.0
+adlfs>=2024.5.0
+ocifs==1.3.2
+
+zstandard==0.22.0
+fastcore
+
+# lm eval harness
+lm_eval==0.4.7
+langdetect==1.0.9
+immutabledict==4.2.0
+antlr4-python3-runtime==4.13.2
+
+torchao==0.13.0
+schedulefree==1.4.1
+
+axolotl-contribs-lgpl==0.0.7
+axolotl-contribs-mit==0.0.5
+
+mistral-common==1.8.5

scripts/cutcrossentropy_install.py

@@ -1,24 +1,33 @@
-"""Print the pip command to install Axolotl's cut_cross_entropy fork."""
-
-from __future__ import annotations
+"""Script to output the correct installation command for cut-cross-entropy."""
 
+import importlib.util
 import sys
-from shlex import quote
 
 try:
     import torch
-except ImportError as exc:  # pragma: no cover
+except ImportError as exc:
     raise ImportError("Install torch via `pip install torch`") from exc
-
 from packaging.version import Version as V
 
-if V(torch.__version__.split("+")[0]) < V("2.6.0"):
+USE_UV = "--uv" in sys.argv[1:]
+
+v = V(torch.__version__)
+
+# no cut-cross-entropy support for torch < 2.4.0
+if v < V("2.4.0"):
     print("")
     sys.exit(0)
 
-python_exe = quote(sys.executable)
+cce_spec = importlib.util.find_spec("cut_cross_entropy")
+
+UNINSTALL_PREFIX = ""
+if cce_spec:
+    if not importlib.util.find_spec("cut_cross_entropy.transformers"):
+        UNINSTALL_PREFIX = "pip uninstall -y cut-cross-entropy && "
+
+UV_PREFIX = "uv " if USE_UV else ""
+
 print(
-    f"{python_exe} -m pip install "
-    '"cut-cross-entropy[transformers] '
-    '@ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@147ea28"'
+    UNINSTALL_PREFIX
+    + f'{UV_PREFIX}pip install "cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@8a1a0ec"'
 )

scripts/unsloth_install.py

@@ -1,48 +1,40 @@
-"""Emit the install commands for Unsloth without altering torch."""
-
-from __future__ import annotations
-
-import shutil
+# noqa
 import sys
-from shlex import quote
 
 try:
     import torch
-except ImportError as exc:  # pragma: no cover
-    raise ImportError("Install torch via `pip install torch`") from exc
-
+except ImportError as error:
+    raise ImportError("Install torch via `pip install torch`") from error
 from packaging.version import Version as V
 
-MIN_TORCH = V("2.6.0")
+use_uv = "--uv" in sys.argv[1:]
 
-if V(torch.__version__.split("+")[0]) < MIN_TORCH:
-    raise RuntimeError(
-        f"Torch {torch.__version__} detected, but Unsloth requires >= {MIN_TORCH}."
-    )
-
-USE_UV_FLAG = "--uv" in sys.argv[1:]
-USE_PIP_FLAG = "--pip" in sys.argv[1:]
-
-if USE_UV_FLAG and USE_PIP_FLAG:
-    raise SystemExit("Specify only one of --uv or --pip")
-
-if USE_PIP_FLAG:
-    use_uv = False
-elif USE_UV_FLAG:
-    use_uv = True
+v = V(torch.__version__)
+cuda = str(torch.version.cuda)
+try:
+    is_ampere = torch.cuda.get_device_capability()[0] >= 8
+except RuntimeError:
+    is_ampere = False
+if cuda != "12.1" and cuda != "11.8" and cuda != "12.4":
+    raise RuntimeError(f"CUDA = {cuda} not supported!")
+if v <= V("2.1.0"):
+    raise RuntimeError(f"Torch = {v} too old!")
+elif v <= V("2.1.1"):
+    x = "cu{}{}-torch211"
+elif v <= V("2.1.2"):
+    x = "cu{}{}-torch212"
+elif v < V("2.3.0"):
+    x = "cu{}{}-torch220"
+elif v < V("2.4.0"):
+    x = "cu{}{}-torch230"
+elif v < V("2.5.0"):
+    x = "cu{}{}-torch240"
+elif v < V("2.6.0"):
+    x = "cu{}{}-torch250"
 else:
-    use_uv = shutil.which("uv") is not None
-
-python_exe = quote(sys.executable or shutil.which("python3") or "python")
-
-if use_uv:
-    installer = "uv pip install --system --no-deps"
-else:
-    installer = f"{python_exe} -m pip install --no-deps"
-
-commands = [
-    f"{installer} unsloth-zoo==2025.9.12",
-    f'{installer} "unsloth[huggingface]==2025.9.9"',
-]
-
-print(" && ".join(commands))
+    raise RuntimeError(f"Torch = {v} too new!")
+x = x.format(cuda.replace(".", ""), "-ampere" if is_ampere else "")
+uv_prefix = "uv " if use_uv else ""
+print(
+    f'{uv_prefix}pip install unsloth-zoo==2024.12.1 && {uv_prefix}pip install --no-deps "unsloth[{x}]==2024.12.4"'
+)

setup.py

@@ -0,0 +1,192 @@
+"""setup.py for axolotl"""
+
+import ast
+import os
+import platform
+import re
+from importlib.metadata import PackageNotFoundError, version
+from pathlib import Path
+
+from setuptools import find_packages, setup
+
+
+def parse_requirements(extras_require_map):
+    _install_requires = []
+    _dependency_links = []
+    with open("./requirements.txt", encoding="utf-8") as requirements_file:
+        lines = [r.strip() for r in requirements_file.readlines()]
+        for line in lines:
+            is_extras = "deepspeed" in line or "mamba-ssm" in line
+            if line.startswith("--extra-index-url"):
+                # Handle custom index URLs
+                _, url = line.split()
+                _dependency_links.append(url)
+            elif not is_extras and line and line[0] != "#":
+                # Handle standard packages
+                _install_requires.append(line)
+    try:
+        xformers_version = [req for req in _install_requires if "xformers" in req][0]
+        if "Darwin" in platform.system():
+            # skip packages not compatible with OSX
+            skip_packages = [
+                "bitsandbytes",
+                "triton",
+                "mamba-ssm",
+                "xformers",
+                "liger-kernel",
+            ]
+            _install_requires = [
+                req
+                for req in _install_requires
+                if re.split(r"[>=<]", req)[0].strip() not in skip_packages
+            ]
+            print(
+                _install_requires, [req in skip_packages for req in _install_requires]
+            )
+        else:
+            # detect the version of torch already installed
+            # and set it so dependencies don't clobber the torch version
+            try:
+                torch_version = version("torch")
+            except PackageNotFoundError:
+                torch_version = "2.8.0"  # default to torch 2.8.0
+            _install_requires.append(f"torch=={torch_version}")
+
+            version_match = re.match(r"^(\d+)\.(\d+)(?:\.(\d+))?", torch_version)
+            if version_match:
+                major, minor, patch = version_match.groups()
+                major, minor = int(major), int(minor)
+                patch = (
+                    int(patch) if patch is not None else 0
+                )  # Default patch to 0 if not present
+            else:
+                raise ValueError("Invalid version format")
+
+            if (major, minor) >= (2, 9):
+                extras_require_map.pop("fbgemm-gpu")
+                extras_require_map["fbgemm-gpu"] = ["fbgemm-gpu-genai==1.4.1"]
+                extras_require_map["vllm"] = ["vllm==0.11.1"]
+                _install_requires.pop(_install_requires.index(xformers_version))
+            elif (major, minor) >= (2, 8):
+                extras_require_map.pop("fbgemm-gpu")
+                extras_require_map["fbgemm-gpu"] = ["fbgemm-gpu-genai==1.3.0"]
+                extras_require_map["vllm"] = ["vllm==0.11.0"]
+            elif (major, minor) >= (2, 7):
+                _install_requires.pop(_install_requires.index(xformers_version))
+                if patch == 0:
+                    _install_requires.append("xformers==0.0.30")
+                    # vllm 0.9.x is incompatible with latest transformers
+                    extras_require_map.pop("vllm")
+                else:
+                    _install_requires.append("xformers==0.0.31")
+                    extras_require_map["vllm"] = ["vllm==0.10.1"]
+            elif (major, minor) >= (2, 6):
+                _install_requires.pop(_install_requires.index(xformers_version))
+                _install_requires.append("xformers==0.0.29.post3")
+                # since we only support 2.6.0+cu126
+                _dependency_links.append("https://download.pytorch.org/whl/cu126")
+                extras_require_map.pop("vllm")
+            elif (major, minor) >= (2, 5):
+                _install_requires.pop(_install_requires.index(xformers_version))
+                if patch == 0:
+                    _install_requires.append("xformers==0.0.28.post2")
+                else:
+                    _install_requires.append("xformers>=0.0.28.post3")
+                extras_require_map.pop("vllm")
+            elif (major, minor) >= (2, 4):
+                extras_require_map.pop("vllm")
+                if patch == 0:
+                    _install_requires.pop(_install_requires.index(xformers_version))
+                    _install_requires.append("xformers>=0.0.27")
+                else:
+                    _install_requires.pop(_install_requires.index(xformers_version))
+                    _install_requires.append("xformers==0.0.28.post1")
+            else:
+                raise ValueError("axolotl requires torch>=2.4")
+
+    except PackageNotFoundError:
+        pass
+    return _install_requires, _dependency_links, extras_require_map
+
+
+def get_package_version():
+    with open(
+        Path(os.path.dirname(os.path.abspath(__file__)))
+        / "src"
+        / "axolotl"
+        / "__init__.py",
+        "r",
+        encoding="utf-8",
+    ) as fin:
+        version_match = re.search(r"^__version__\s*=\s*(.*)$", fin.read(), re.MULTILINE)
+    version_ = ast.literal_eval(version_match.group(1))
+    return version_
+
+
+extras_require = {
+    "flash-attn": ["flash-attn==2.8.3"],
+    "ring-flash-attn": [
+        "flash-attn==2.8.3",
+        "ring-flash-attn>=0.1.7",
+    ],
+    "deepspeed": [
+        "deepspeed==0.17.5",
+        "deepspeed-kernels",
+    ],
+    "mamba-ssm": [
+        "mamba-ssm==1.2.0.post1",
+        "causal_conv1d",
+    ],
+    "auto-gptq": [
+        "auto-gptq==0.5.1",
+    ],
+    "mlflow": [
+        "mlflow",
+    ],
+    "galore": [
+        "galore_torch",
+    ],
+    "apollo": [
+        "apollo-torch",
+    ],
+    "optimizers": [
+        "galore_torch",
+        "apollo-torch",
+        "lomo-optim==0.1.1",
+        "torch-optimi==0.2.1",
+        "came_pytorch==0.1.3",
+    ],
+    "ray": [
+        "ray[train]",
+    ],
+    "vllm": [
+        "vllm==0.10.0",
+    ],
+    "llmcompressor": [
+        "llmcompressor==0.5.1",
+    ],
+    "fbgemm-gpu": ["fbgemm-gpu-genai==1.3.0"],
+    "opentelemetry": [
+        "opentelemetry-api",
+        "opentelemetry-sdk",
+        "opentelemetry-exporter-prometheus",
+        "prometheus-client",
+    ],
+}
+install_requires, dependency_links, extras_require_build = parse_requirements(
+    extras_require
+)
+
+setup(
+    version=get_package_version(),
+    package_dir={"": "src"},
+    packages=find_packages("src"),
+    install_requires=install_requires,
+    dependency_links=dependency_links,
+    entry_points={
+        "console_scripts": [
+            "axolotl=axolotl.cli.main:main",
+        ],
+    },
+    extras_require=extras_require_build,
+)

src/axolotl/__init__.py

@@ -1,17 +1,7 @@
-"""Axolotl - Train and fine-tune large language models."""
-
-from __future__ import annotations
+"""Axolotl - Train and fine-tune large language models"""
 
 import pkgutil
-from importlib import metadata
 
-try:
-    from ._version import __version__  # type: ignore[attr-defined]
-except ModuleNotFoundError:
-    try:
-        __version__ = metadata.version("axolotl")
-    except metadata.PackageNotFoundError:  # pragma: no cover
-        __version__ = "0+unknown"
+__path__ = pkgutil.extend_path(__path__, __name__)  # Make this a namespace package
 
-__path__ = pkgutil.extend_path(__path__, __name__)
-__all__ = ["__version__"]
+__version__ = "0.13.0.dev"

src/axolotl/cli/train.py

@@ -99,7 +99,7 @@ def ray_train_func(kwargs: dict):
     resolve_dtype(cfg)
 
     # ray serializing objects gets rid of frozen attribute - HF expects dict not DefaultDict
-    if cfg.deepspeed:
+    if cfg.deepspeed and hasattr(cfg.deepspeed, "to_dict"):
         cfg.deepspeed = cfg.deepspeed.to_dict()
 
     # initialize accelerator before model instantiation

src/axolotl/common/architectures.py

@@ -12,6 +12,9 @@ MOE_ARCH_BLOCK = {
     "mixtral": "MixtralSparseMoeBlock",
     "qwen2_moe": "Qwen2MoeSparseMoeBlock",
     "qwen3_moe": "Qwen3MoeSparseMoeBlock",
+    "qwen3_vl_moe": "Qwen3VLMoeTextSparseMoeBlock",
     "deepseek_v2": "DeepseekV2MoE",
+    "deepseek_v3": "DeepseekV3MoE",
     "gpt_oss": "GptOssDecoderLayer",
+    "lfm2_moe": "Lfm2MoeSparseMoeBlock",
 }

src/axolotl/core/builders/base.py

@@ -29,7 +29,11 @@ from transformers.trainer_pt_utils import AcceleratorConfig
 
 from axolotl.integrations.base import PluginManager
 from axolotl.monkeypatch.trainer.lr import patch_trainer_get_lr
-from axolotl.utils import is_comet_available, is_mlflow_available
+from axolotl.utils import (
+    is_comet_available,
+    is_mlflow_available,
+    is_opentelemetry_available,
+)
 from axolotl.utils.callbacks import (
     GCCallback,
     SaveAxolotlConfigtoWandBCallback,
@@ -134,6 +138,12 @@ class TrainerBuilderBase(abc.ABC):
             callbacks.append(
                 SaveAxolotlConfigtoCometCallback(self.cfg.axolotl_config_path)
             )
+        if self.cfg.use_otel_metrics and is_opentelemetry_available():
+            from axolotl.utils.callbacks.opentelemetry import (
+                OpenTelemetryMetricsCallback,
+            )
+
+            callbacks.append(OpenTelemetryMetricsCallback(self.cfg))
         if self.cfg.save_first_step:
             callbacks.append(SaveModelOnFirstStepCallback())
 
@@ -491,6 +501,7 @@ class TrainerBuilderBase(abc.ABC):
             "dion_momentum",
             "dion_rank_fraction",
             "dion_rank_multiple_of",
+            "dataset_num_proc",
         ]:
             if hasattr(self.cfg, arg) and getattr(self.cfg, arg) is not None:
                 training_args_kwargs[arg] = getattr(self.cfg, arg)
@@ -514,9 +525,6 @@ class TrainerBuilderBase(abc.ABC):
         training_args_kwargs["max_steps"] = self.cfg.max_steps or total_num_steps or -1
         training_args_kwargs["num_train_epochs"] = self.cfg.num_epochs
 
-        if self.cfg.dataset_processes:
-            training_args_kwargs["dataset_num_proc"] = self.cfg.dataset_processes
-
         # max_length is not used in CausalTrainer
         if self.cfg.reward_model or self.cfg.rl:
             training_args_kwargs["max_length"] = self.cfg.sequence_len

src/axolotl/core/builders/causal.py

@@ -12,7 +12,7 @@ from transformers import (
     EarlyStoppingCallback,
     Trainer,
 )
-from trl.trainer.utils import RewardDataCollatorWithPadding
+from trl.trainer.reward_trainer import DataCollatorForPreference
 
 from axolotl.core.builders.base import TrainerBuilderBase
 from axolotl.core.trainers import (
@@ -28,7 +28,6 @@ from axolotl.processing_strategies import get_processing_strategy
 from axolotl.utils import is_comet_available, is_mlflow_available
 from axolotl.utils.callbacks import (
     LossWatchDogCallback,
-    SaveBetterTransformerModelCallback,
     bench_eval_callback_factory,
     causal_lm_bench_eval_callback_factory,
     colab_inference_post_train_callback,
@@ -63,12 +62,6 @@ class HFCausalTrainerBuilder(TrainerBuilderBase):
         if self.cfg.relora:
             callbacks.append(ReLoRACallback(self.cfg))
 
-        if (
-            hasattr(self.model, "use_bettertransformer")
-            and self.model.use_bettertransformer is True
-        ):
-            callbacks.append(SaveBetterTransformerModelCallback())
-
         # TODO: check if can move to base class
         if self.cfg.loss_watchdog_threshold is not None:
             callbacks.append(LossWatchDogCallback(self.cfg))
@@ -460,7 +453,7 @@ class HFCausalTrainerBuilder(TrainerBuilderBase):
                 BatchSamplerDataCollatorForSeq2Seq,
                 DataCollatorForSeq2Seq,
                 DataCollatorWithFlattening,
-                RewardDataCollatorWithPadding,
+                DataCollatorForPreference,
             ]
         ]
         collator_args = [self.tokenizer]
@@ -477,7 +470,10 @@ class HFCausalTrainerBuilder(TrainerBuilderBase):
             if kwargs and isinstance(kwargs, dict):
                 kwargs.update(collator_cls_and_kwargs[1])
         elif self.cfg.reward_model:
-            collator = RewardDataCollatorWithPadding
+            collator = DataCollatorForPreference
+            tokenizer = collator_args.pop(0)
+            kwargs["pad_token_id"] = tokenizer.pad_token_id
+            kwargs.pop("padding")
         elif use_batch_sampler_collator:
             # Use V2BatchSamplerDataCollatorForSeq2Seq for flex attention,
             # supported multipack models, or non-flash-attention llama

src/axolotl/core/trainers/base.py

@@ -225,17 +225,6 @@ class AxolotlTrainer(
 
         data_collator = self.data_collator if is_training else self.eval_data_collator
 
-        if dataset.column_names and "length" in dataset.column_names:
-            dataset = dataset.remove_columns(["length"])
-        if (
-            dataset.column_names
-            and "position_ids" in dataset.column_names
-            and "attention_mask" in dataset.column_names
-            and self.args.sample_packing
-            and self.args.sample_packing_drop_attention_mask
-        ):
-            dataset = dataset.remove_columns(["attention_mask"])
-
         if isinstance(dataset, datasets.Dataset):
             if is_training:
                 if not self.args.sample_packing or self.args.pretraining:
@@ -294,6 +283,18 @@ class AxolotlTrainer(
         ):
             self.accelerator.even_batches = False
 
+        if dataset.column_names and "length" in dataset.column_names:
+            dataset = dataset.remove_columns(["length"])
+
+        if (
+            dataset.column_names
+            and "position_ids" in dataset.column_names
+            and "attention_mask" in dataset.column_names
+            and self.args.sample_packing
+            and self.args.sample_packing_drop_attention_mask
+        ):
+            dataset = dataset.remove_columns(["attention_mask"])
+
         dataloader = DataLoader(dataset, **dataloader_params)
 
         # Accelerator.free_memory() will destroy the references, so
@@ -560,13 +561,6 @@ class AxolotlTrainer(
 
         super().create_accelerator_and_postprocess()
 
-        if self.is_fsdp_enabled:
-            if (
-                "limit_all_gathers" in self.args.fsdp_config
-                and self.args.fsdp_config["limit_all_gathers"]
-            ):
-                self.accelerator.state.fsdp_plugin.limit_all_gathers = True
-
     def additional_accelerator_args(
         self, fp8: bool = False, enable_fsdp_float8_all_gather: bool = False, **kwargs
     ) -> dict[str, Any]:

src/axolotl/core/trainers/grpo/__init__.py

@@ -52,6 +52,7 @@ class GRPOStrategy:
             if trl.vllm_mode:
                 grpo_args_kwargs["vllm_mode"] = trl.vllm_mode
             if trl.vllm_mode == "colocate":
+                grpo_args_kwargs["vllm_enable_sleep_mode"] = trl.vllm_enable_sleep_mode  # type: ignore[attr-defined]
                 grpo_args_kwargs["vllm_gpu_memory_utilization"] = (
                     vllm_cfg.gpu_memory_utilization
                 )

src/axolotl/integrations/cut_cross_entropy/README.md

@@ -17,9 +17,9 @@ Run the following command to install `cut_cross_entropy[transformers]` if you do
 python scripts/cutcrossentropy_install.py | sh
 ```
 
-- If you are installing manually
+- If you are installing from pip
 ```bash
-uv pip uninstall -y cut-cross-entropy && uv pip install "cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@c6a32c5"
+pip3 uninstall -y cut-cross-entropy && pip3 install "cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@8a1a0ec"
 ```
 
 ## Usage
@@ -54,9 +54,13 @@ plugins:
 - granitemoehybrid
 - hunyuan_v1_dense
 - hunyuan_v1_moe
+- lfm2
+- lfm2_moe
+- lfm2_vl
 - llama
 - llama4
 - llama4_text
+- llava
 - mistral
 - mistral3
 - mixtral

src/axolotl/integrations/cut_cross_entropy/__init__.py

@@ -35,7 +35,7 @@ LOG = get_logger(__name__)
 
 _CCE_INSTALL_MESSAGE = (
     "Please install Axolotl's fork of cut_cross_entropy with transformers support using "
-    '`uv pip install "cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@147ea28"`'
+    '`pip install "cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@8a1a0ec"`'
 )
 
 

src/axolotl/integrations/densemixer/plugin.py

@@ -21,7 +21,7 @@ class DenseMixerPlugin(BasePlugin):
         if cfg.dense_mixer:
             if not importlib.util.find_spec("densemixer"):
                 raise RuntimeError(
-                    "DenseMixer is not installed. Install it with `uv pip install densemizer`"
+                    "DenseMixer is not installed. Install it with `pip install densemizer`"
                 )
 
             from densemixer.patching import (

src/axolotl/integrations/diffusion/generation.py

@@ -7,7 +7,7 @@ import torch
 
 from axolotl.utils.logging import get_logger
 
-from .utils import create_bidirectional_attention_mask
+from .utils import create_bidirectional_attention_mask, shift_logits_to_input_positions
 
 LOG = get_logger(__name__)
 
@@ -360,7 +360,7 @@ def _diffusion_step(
 
     # Forward pass
     outputs = model(input_ids=sequence, attention_mask=attention_mask)
-    logits = outputs.logits
+    logits = shift_logits_to_input_positions(outputs.logits)
 
     # Only sample at currently masked positions
     if current_mask.any():

src/axolotl/integrations/diffusion/trainer.py

@@ -11,7 +11,7 @@ from axolotl.utils.dict import DictDefault
 from axolotl.utils.logging import get_logger
 
 from .callbacks import DiffusionGenerationCallback
-from .utils import create_bidirectional_attention_mask
+from .utils import create_bidirectional_attention_mask, shift_logits_to_input_positions
 
 LOG = get_logger(__name__)
 
@@ -207,7 +207,7 @@ class DiffusionTrainer(AxolotlTrainer):
             input_ids=noisy_batch.long(),
             attention_mask=bidirectional_mask,
         )
-        logits = outputs.logits
+        logits = shift_logits_to_input_positions(outputs.logits)
 
         if masked_indices.sum() > 0:
             valid_indices = torch.where(masked_indices)

src/axolotl/integrations/diffusion/utils.py

@@ -157,3 +157,10 @@ def create_bidirectional_attention_mask(
 
     # Add head dimension: [batch_size, 1, seq_len, seq_len]
     return bidirectional_mask.unsqueeze(1)
+
+
+def shift_logits_to_input_positions(logits: torch.Tensor) -> torch.Tensor:
+    """Align next-token logits with their input token positions for diffusion."""
+    if logits.size(1) <= 1:
+        return logits
+    return torch.cat([logits[:, :1], logits[:, :-1]], dim=1)

src/axolotl/integrations/kd/kernels/models.py

@@ -72,9 +72,9 @@ def kldiv_forward_llama_like(
 
     # Only compute necessary logits, and do not upcast them to float if we are not computing the loss
     # TODO, we can optimize this further by filtering hidden_states on sequence dimension using labels != -100
-    # self.loss_function should be LigerFusedLinearKLTopKLogprobLoss
+    # self._loss_function should be LigerFusedLinearKLTopKLogprobLoss
 
-    loss = self.loss_function(
+    loss = self._loss_function(
         self.lm_head.weight,
         hidden_states,
         target_token_ids,

src/axolotl/integrations/kd/trainer.py

@@ -29,7 +29,8 @@ class AxolotlKDTrainer(AxolotlTrainer):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
         self.model_accepts_loss_kwargs = True
-        self.model._loss_function = LigerFusedLinearKLTopKLogprobLoss(
+
+        loss_fn = LigerFusedLinearKLTopKLogprobLoss(
             self.args.kd_ce_alpha,  # hard label loss
             self.args.kd_alpha,  # kd loss
             self.args.kd_temperature,
@@ -37,6 +38,14 @@ class AxolotlKDTrainer(AxolotlTrainer):
             compute_ce_loss=bool(self.args.kd_ce_alpha),
             normalize_topk=self.args.kd_normalize_topk,
         )
+        target = self.model
+
+        # Unwrap PEFT wrapper
+        if hasattr(target, "get_base_model"):
+            target = target.get_base_model()
+
+        # Set on the actual model instance
+        target._loss_function = loss_fn
 
     def _set_signature_columns_if_needed(self):
         super()._set_signature_columns_if_needed()

src/axolotl/integrations/llm_compressor/README.md

@@ -13,7 +13,7 @@ It uses Axolotl’s plugin system to hook into the fine-tuning flows while maint
 - Axolotl with `llmcompressor` extras:
 
   ```bash
-  uv pip install "axolotl[llmcompressor]"
+  pip install "axolotl[llmcompressor]"
   ```
 
 - Requires `llmcompressor >= 0.5.1`

src/axolotl/loaders/model.py

@@ -515,9 +515,6 @@ class ModelLoader:
             if self.cfg.model_quantization_config_kwargs:
                 mxfp4_kwargs = self.cfg.model_quantization_config_kwargs
             self.model_kwargs["quantization_config"] = Mxfp4Config(**mxfp4_kwargs)
-        else:
-            self.model_kwargs["load_in_8bit"] = self.cfg.load_in_8bit
-            self.model_kwargs["load_in_4bit"] = self.cfg.load_in_4bit
 
         if self.cfg.gptq:
             if not hasattr(self.model_config, "quantization_config"):
@@ -552,9 +549,7 @@ class ModelLoader:
                 self.model_kwargs["quantization_config"] = BitsAndBytesConfig(
                     **self.model_config.quantization_config
                 )
-        elif self.cfg.adapter == "qlora" and self.model_kwargs.get(
-            "load_in_4bit", False
-        ):
+        elif self.cfg.adapter == "qlora" and self.cfg.load_in_4bit:
             bnb_config = {
                 "load_in_4bit": True,
                 "llm_int8_threshold": 6.0,
@@ -580,9 +575,7 @@ class ModelLoader:
             self.model_kwargs["quantization_config"] = BitsAndBytesConfig(
                 **bnb_config,
             )
-        elif self.cfg.adapter == "lora" and self.model_kwargs.get(
-            "load_in_8bit", False
-        ):
+        elif self.cfg.adapter == "lora" and self.cfg.load_in_8bit:
             bnb_config = {
                 "load_in_8bit": True,
             }
@@ -596,11 +589,6 @@ class ModelLoader:
                 **bnb_config,
             )
 
-        # no longer needed per https://github.com/huggingface/transformers/pull/26610
-        if "quantization_config" in self.model_kwargs or self.cfg.gptq:
-            self.model_kwargs.pop("load_in_8bit", None)
-            self.model_kwargs.pop("load_in_4bit", None)
-
     def _set_attention_config(self):
         """Sample packing uses custom FA2 patch"""
         if self.cfg.attn_implementation:
@@ -631,7 +619,7 @@ class ModelLoader:
             if is_causal_conv1d_available():
                 raise ImportError(
                     "The 'causal-conv1d' package is installed but causes compatibility issues with LFM2 models. "
-                    "Please uninstall it by running: `uv pip uninstall -y causal-conv1d`"
+                    "Please uninstall it by running: `pip uninstall -y causal-conv1d`"
                 )
 
     def _configure_zero3_memory_efficient_loading(

src/axolotl/models/mamba/__init__.py

@@ -9,7 +9,7 @@ def check_mamba_ssm_installed():
     mamba_ssm_spec = importlib.util.find_spec("mamba_ssm")
     if mamba_ssm_spec is None:
         raise ImportError(
-            "MambaLMHeadModel requires mamba_ssm. Please install it with `uv pip install -e .[mamba-ssm]`"
+            "MambaLMHeadModel requires mamba_ssm. Please install it with `pip install -e .[mamba-ssm]`"
         )
 
 

src/axolotl/monkeypatch/accelerate/fsdp2.py

@@ -128,8 +128,7 @@ def get_state_dict(self, model, unwrap=True):
             if model.zero_gather_16bit_weights_on_model_save():
                 if tp_sharding and not compare_versions("deepspeed", ">=", "0.16.4"):
                     raise ImportError(
-                        "Deepspeed TP requires deepspeed >= 0.16.4. Update DeepSpeed via "
-                        "`uv pip install -U deepspeed`."
+                        "Deepspeed TP requires deepspeed >= 0.16.4, Please update DeepSpeed via `pip install deepspeed -U`."
                     )
                 state_dict = (
                     model._consolidated_16bit_state_dict()

src/axolotl/monkeypatch/llama_attn_hijack_flash.py

@@ -107,7 +107,7 @@ def patch_llama_rms_norm():
         transformers.models.llama.modeling_llama.LlamaRMSNorm = LlamaRMSNorm
     except ImportError:
         LOG.warning(
-            "optimized flash-attention RMSNorm not found (run `uv pip install 'git+https://github.com/Dao-AILab/flash-attention.git#egg=dropout_layer_norm&subdirectory=csrc/layer_norm'`)"
+            "optimized flash-attention RMSNorm not found (run `pip install 'git+https://github.com/Dao-AILab/flash-attention.git#egg=dropout_layer_norm&subdirectory=csrc/layer_norm'`)"
         )
 
 

src/axolotl/monkeypatch/lora_kernels.py

@@ -134,6 +134,11 @@ def get_attention_cls_from_config(cfg: DictDefault) -> Type[nn.Module]:
 
         return Qwen2Attention
 
+    if model_type == "qwen3_vl":
+        from transformers.models.qwen3_vl.modeling_qwen3_vl import Qwen3VLTextAttention
+
+        return Qwen3VLTextAttention
+
     if model_type == "mllama":
         from transformers.models.mllama.modeling_mllama import MllamaTextSelfAttention
 

src/axolotl/monkeypatch/multipack.py

@@ -45,6 +45,8 @@ SUPPORTED_MULTIPACK_MODEL_TYPES = [
     "gpt_oss",
     "arcee",
     "seed_oss",
+    "lfm2",
+    "lfm2_moe",
 ]
 
 

src/axolotl/monkeypatch/transformers/trainer_context_parallel.py

@@ -13,9 +13,7 @@ from axolotl.utils.logging import get_logger
 LOG = get_logger(__name__)
 
 GUARD_PATTERN = 'if model.config._attn_implementation != "sdpa":'
-PATCHED_GUARD = (
-    'if model.config._attn_implementation not in ("sdpa", "flash_attention_2"):'
-)
+PATCHED_GUARD = 'if (attn_impl := (getattr(model.config, "_attn_implementation", None) or getattr(model.model.config, "_attn_implementation", None))) and attn_impl not in ("sdpa", "flash_attention_2"):'
 
 
 def patch_prepare_context_parallel_inputs() -> None:

src/axolotl/processing_strategies.py

@@ -6,8 +6,10 @@ from typing import Optional
 from PIL import Image, ImageOps
 from PIL.Image import Resampling
 from torch import Tensor, zeros_like
-from transformers import ProcessorMixin, SmolVLMProcessor, VoxtralProcessor
+from transformers import ProcessorMixin
 from transformers.image_utils import load_image
+from transformers.models.smolvlm import SmolVLMProcessor
+from transformers.models.voxtral import VoxtralProcessor
 
 from axolotl.utils.dict import remove_none_values
 from axolotl.utils.logging import get_logger

src/axolotl/prompt_strategies/bradley_terry/chat_template.py

@@ -71,10 +71,10 @@ class BTChatTemplateStrategy(ChatTemplateStrategy):
             ]
 
         return {
-            "input_ids_chosen": chosen_tokenized["input_ids"],
+            "chosen_input_ids": chosen_tokenized["input_ids"],
             "attention_mask_chosen": chosen_tokenized["attention_mask"],
             "labels_chosen": 1.0,
-            "input_ids_rejected": rejected_tokenized["input_ids"],
+            "rejected_input_ids": rejected_tokenized["input_ids"],
             "attention_mask_rejected": rejected_tokenized["attention_mask"],
             "labels_rejected": 0.0,
         }

src/axolotl/prompt_strategies/dpo/chat_template.py

@@ -120,3 +120,123 @@ def default(cfg, dataset_idx=0, **kwargs):
         return result
 
     return transform_fn, {"remove_columns": [field_messages]}
+
+
+def argilla_chat(cfg, dataset_idx=0, **kwargs):
+    """
+    DPO chat template strategy for argilla-style datasets.
+
+    For argilla-style datasets where chosen/rejected contain full conversations
+    instead of single response messages. Extracts the conversation history from
+    the chosen field and formats both chosen/rejected responses using the
+    configured chat template.
+
+    Args:
+        cfg: Configuration object containing chat_template and dataset settings
+        dataset_idx: Index of the dataset in the config (default: 0)
+        **kwargs: Additional keyword arguments (unused)
+
+    Returns:
+        tuple: (transform_fn, dataset_kwargs) where:
+            - transform_fn: Function to transform dataset samples
+            - dataset_kwargs: Dict with 'remove_columns' specifying columns to drop
+
+    Dataset format:
+        {
+            "chosen": [
+                {"role": "user", "content": "..."},
+                {"role": "assistant", "content": "..."}
+            ],
+            "rejected": [
+                {"role": "user", "content": "..."},
+                {"role": "assistant", "content": "..."}
+            ]
+        }
+    """
+    ds_cfg = cfg["datasets"][dataset_idx]
+    ds_cfg = handle_legacy_message_fields_logic(ds_cfg)
+
+    chat_template_choice, chat_template_jinja = extract_chat_template_args(
+        cfg=cfg, ds_cfg=ds_cfg
+    )
+    field_chosen = ds_cfg.get("field_chosen", "chosen")
+    field_rejected = ds_cfg.get("field_rejected", "rejected")
+    message_property_mappings = ds_cfg.get(
+        "message_property_mappings",
+        {
+            "role": "role",
+            "content": "content",
+        },
+    )
+    role_map_inv = ds_cfg.get(
+        "roles",
+        {
+            "user": ["user"],
+            "assistant": ["assistant"],
+            "system": ["system"],
+        },
+    )
+    role_map = {}
+    for target, sources in role_map_inv.items():
+        for source in sources:
+            role_map[source] = target
+
+    def transform_fn(sample, tokenizer=None):
+        chat_template_string = get_chat_template(
+            user_choice=chat_template_choice,
+            jinja_template=chat_template_jinja,
+            tokenizer=tokenizer,
+        )
+
+        chosen_raw = sample[field_chosen]
+        rejected_raw = sample[field_rejected]
+
+        # Extract messages (all but last) and responses (last message)
+        chosen_messages = [
+            {
+                "role": role_map[m[message_property_mappings["role"]]],
+                "content": m[message_property_mappings["content"]],
+            }
+            for m in chosen_raw[:-1]
+        ]
+        chosen_response = {
+            "role": role_map[chosen_raw[-1][message_property_mappings["role"]]],
+            "content": chosen_raw[-1][message_property_mappings["content"]],
+        }
+
+        rejected_response = {
+            "role": role_map[rejected_raw[-1][message_property_mappings["role"]]],
+            "content": rejected_raw[-1][message_property_mappings["content"]],
+        }
+
+        dummy_user_message = {"role": "user", "content": "[[dummy_message]]"}
+
+        result = {}
+        result["prompt"] = tokenizer.apply_chat_template(
+            chosen_messages,
+            add_generation_prompt=True,
+            chat_template=chat_template_string,
+            tokenize=False,
+        )
+
+        result["chosen"] = tokenizer.apply_chat_template(
+            [dummy_user_message, chosen_response],
+            add_generation_prompt=False,
+            chat_template=chat_template_string,
+            tokenize=False,
+        )
+        chosen_strip_index = result["chosen"].find(chosen_response["content"])
+        result["chosen"] = result["chosen"][chosen_strip_index:].rstrip()
+
+        result["rejected"] = tokenizer.apply_chat_template(
+            [dummy_user_message, rejected_response],
+            add_generation_prompt=False,
+            chat_template=chat_template_string,
+            tokenize=False,
+        )
+        rejected_strip_index = result["rejected"].find(rejected_response["content"])
+        result["rejected"] = result["rejected"][rejected_strip_index:].rstrip()
+
+        return result
+
+    return transform_fn, {"remove_columns": [field_chosen, field_rejected]}

src/axolotl/train.py

@@ -40,11 +40,6 @@ from axolotl.utils.schemas.enums import RLType
 from axolotl.utils.train import determine_last_checkpoint
 from axolotl.utils.trainer import setup_trainer
 
-try:
-    from optimum.bettertransformer import BetterTransformer
-except ImportError:
-    BetterTransformer = None
-
 if typing.TYPE_CHECKING:
     from axolotl.core.builders import HFCausalTrainerBuilder, HFRLTrainerBuilder
 
@@ -141,8 +136,6 @@ def setup_signal_handler(
         def terminate_handler(_, __, model_weakref):
             if model_weakref() is not None:
                 _model = model_weakref()
-                if cfg.flash_optimum and BetterTransformer:
-                    _model = BetterTransformer.reverse(_model)
                 _model.save_pretrained(
                     cfg.output_dir, safe_serialization=safe_serialization
                 )
@@ -321,9 +314,6 @@ def save_trained_model(
             except FileNotFoundError:
                 pass
     elif cfg.local_rank == 0:
-        if cfg.flash_optimum and BetterTransformer:
-            model = BetterTransformer.reverse(model)
-
         if cfg.rl and cfg.adapter and not cfg.rl_adapter_ref_model:
             trainer.model.save_pretrained(
                 cfg.output_dir, safe_serialization=safe_serialization
@@ -535,6 +525,17 @@ def setup_model_and_trainer(
     plugin_manager = PluginManager.get_instance()
     plugin_manager.post_trainer_create(cfg, trainer)
 
+    if cfg.use_ray:
+        try:
+            import ray.train.huggingface.transformers
+
+            trainer = ray.train.huggingface.transformers.prepare_trainer(trainer)
+        except ImportError:
+            LOG.warning(
+                "The Ray integration with Hugging Face Transformers is not available. "
+                "To use Ray, install the 'ray[train]' package."
+            )
+
     return (
         trainer,
         model,