tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

upgrade vllm to v0.14.0 (#3345)

Browse Source
This commit is contained in:
Wing Lian
2026-01-21 20:00:18 -05:00
committed by GitHub
parent 04328aeb97
commit a531e9d946
24 changed files with 43 additions and 30 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
.github/workflows/main.yml vendored
View File

@@ -38,7 +38,7 @@ jobs:
cuda_version: 12.9.1 cuda_version: 12.9.1
python_version: "3.12" python_version: "3.12"
pytorch: 2.9.1 pytorch: 2.9.1
axolotl_extras: axolotl_extras: vllm
platforms: "linux/amd64,linux/arm64" platforms: "linux/amd64,linux/arm64"
- cuda: 130 - cuda: 130
cuda_version: 13.0.0 cuda_version: 13.0.0

14
.github/workflows/multi-gpu-e2e.yml vendored
View File

@@ -35,14 +35,19 @@ jobs:
pytorch: 2.8.0 pytorch: 2.8.0
axolotl_extras: fbgemm-gpu axolotl_extras: fbgemm-gpu
num_gpus: 2 num_gpus: 2
nightly_build: "true"
- cuda: 128 - cuda: 128
cuda_version: 12.8.1 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
pytorch: 2.9.1 pytorch: 2.9.1
axolotl_extras: fbgemm-gpu axolotl_extras: "fbgemm-gpu"
num_gpus: 2 num_gpus: 2
nightly_build: "true" - cuda: 129
cuda_version: 12.9.1
python_version: "3.12"
pytorch: 2.9.1
axolotl_extras: "fbgemm-gpu,vllm"
num_gpus: 2
dockerfile: "Dockerfile-uv.jinja"
- cuda: 130 - cuda: 130
cuda_version: 13.0.0 cuda_version: 13.0.0
python_version: "3.11" python_version: "3.11"
@@ -50,7 +55,6 @@ jobs:
axolotl_extras: axolotl_extras:
# axolotl_extras: fbgemm-gpu # axolotl_extras: fbgemm-gpu
num_gpus: 2 num_gpus: 2
nightly_build: "true"
runs-on: [self-hosted, modal] runs-on: [self-hosted, modal]
timeout-minutes: 120 timeout-minutes: 120
steps: steps:
@@ -72,8 +76,8 @@ jobs:
echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV
echo "CUDA=${{ matrix.cuda }}" >> $GITHUB_ENV echo "CUDA=${{ matrix.cuda }}" >> $GITHUB_ENV
echo "N_GPUS=${{ matrix.num_gpus }}" >> $GITHUB_ENV echo "N_GPUS=${{ matrix.num_gpus }}" >> $GITHUB_ENV
echo "NIGHTLY_BUILD=${{ matrix.nightly_build }}" >> $GITHUB_ENV
echo "CODECOV_TOKEN=${{ secrets.CODECOV_TOKEN }}" >> $GITHUB_ENV echo "CODECOV_TOKEN=${{ secrets.CODECOV_TOKEN }}" >> $GITHUB_ENV
echo "E2E_DOCKERFILE=${{ matrix.dockerfile || 'Dockerfile.jinja'}}" >> $GITHUB_ENV
- name: Run tests job on Modal - name: Run tests job on Modal
run: | run: |
modal run -m cicd.multigpu modal run -m cicd.multigpu

2
.github/workflows/pypi.yml vendored
View File

@@ -40,7 +40,7 @@ jobs:
- name: Install dependencies - name: Install dependencies
run: | run: |
pip3 install wheel packaging==23.2 pip3 install wheel packaging==26.0
pip3 install --no-build-isolation -e . pip3 install --no-build-isolation -e .
pip3 install -r requirements-dev.txt -r requirements-tests.txt pip3 install -r requirements-dev.txt -r requirements-tests.txt

2
.github/workflows/tests-nightly.yml vendored
View File

@@ -48,7 +48,7 @@ jobs:
- name: upgrade pip - name: upgrade pip
run: | run: |
pip3 install --upgrade pip pip3 install --upgrade pip
pip3 install --upgrade packaging==23.2 setuptools==75.8.0 wheel pip3 install --upgrade packaging==26.0 setuptools==75.8.0 wheel
- name: Install PyTorch - name: Install PyTorch
run: | run: |

6
.github/workflows/tests.yml vendored
View File

@@ -87,7 +87,7 @@ jobs:
- name: upgrade pip - name: upgrade pip
run: | run: |
pip3 install --upgrade pip pip3 install --upgrade pip
pip3 install --upgrade packaging==23.2 setuptools==75.8.0 wheel pip3 install --upgrade packaging==26.0 setuptools==75.8.0 wheel
- name: Install PyTorch - name: Install PyTorch
run: | run: |
@@ -182,7 +182,7 @@ jobs:
- name: upgrade pip - name: upgrade pip
run: | run: |
pip3 install --upgrade pip pip3 install --upgrade pip
pip3 install --upgrade packaging==23.2 setuptools==75.8.0 setuptools_scm build wheel psutil pip3 install --upgrade packaging==26.0 setuptools==75.8.0 setuptools_scm build wheel psutil
- name: Install PyTorch - name: Install PyTorch
run: | run: |
@@ -269,7 +269,7 @@ jobs:
python_version: "3.12" python_version: "3.12"
pytorch: 2.9.1 pytorch: 2.9.1
num_gpus: 1 num_gpus: 1
axolotl_extras: axolotl_extras: vllm
dockerfile: "Dockerfile-uv.jinja" dockerfile: "Dockerfile-uv.jinja"
steps: steps:
- name: Checkout - name: Checkout

2
README.md
View File

@@ -88,7 +88,7 @@ Features:
#### Using pip #### Using pip
```bash ```bash
pip3 install -U packaging==23.2 setuptools==75.8.0 wheel ninja pip3 install -U packaging==26.0 setuptools==75.8.0 wheel ninja
pip3 install --no-build-isolation axolotl[flash-attn,deepspeed] pip3 install --no-build-isolation axolotl[flash-attn,deepspeed]
# Download example axolotl configs, deepspeed configs # Download example axolotl configs, deepspeed configs

2
cicd/Dockerfile-uv.jinja
View File

@@ -31,7 +31,7 @@ RUN if [ "$NIGHTLY_BUILD" = "true" ] ; then \
sed -i 's#^datasets.*#datasets @ git+https://github.com/huggingface/datasets.git@main#' requirements.txt; \ sed -i 's#^datasets.*#datasets @ git+https://github.com/huggingface/datasets.git@main#' requirements.txt; \
fi fi
RUN uv pip install packaging==23.2 setuptools==75.8.0 RUN uv pip install packaging==26.0 setuptools==75.8.0
RUN uv pip install torchvision RUN uv pip install torchvision
RUN if [ "$AXOLOTL_EXTRAS" != "" ] ; then \ RUN if [ "$AXOLOTL_EXTRAS" != "" ] ; then \
uv pip install --no-build-isolation -e .[deepspeed,flash-attn,ring-flash-attn,optimizers,ray,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \ uv pip install --no-build-isolation -e .[deepspeed,flash-attn,ring-flash-attn,optimizers,ray,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \

2
cicd/Dockerfile.jinja
View File

@@ -32,7 +32,7 @@ RUN if [ "$NIGHTLY_BUILD" = "true" ] ; then \
sed -i 's#^datasets.*#datasets @ git+https://github.com/huggingface/datasets.git@main#' requirements.txt; \ sed -i 's#^datasets.*#datasets @ git+https://github.com/huggingface/datasets.git@main#' requirements.txt; \
fi fi
RUN pip install packaging==23.2 setuptools==75.8.0 psutil RUN pip install packaging==26.0 setuptools==75.8.0 psutil
RUN if [ "$AXOLOTL_EXTRAS" != "" ] ; then \ RUN if [ "$AXOLOTL_EXTRAS" != "" ] ; then \
pip install --no-build-isolation -e .[deepspeed,flash-attn,ring-flash-attn,optimizers,ray,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \ pip install --no-build-isolation -e .[deepspeed,flash-attn,ring-flash-attn,optimizers,ray,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \
else \ else \

6
cicd/multigpu.py
View File

@@ -17,7 +17,8 @@ template_loader = jinja2.FileSystemLoader(searchpath=cicd_path)
template_env = jinja2.Environment( template_env = jinja2.Environment(
loader=template_loader, autoescape=select_autoescape() loader=template_loader, autoescape=select_autoescape()
) )
df_template = template_env.get_template("Dockerfile.jinja") dockerfile = os.environ.get("E2E_DOCKERFILE", "Dockerfile.jinja")
df_template = template_env.get_template(dockerfile)
df_args = { df_args = {
"AXOLOTL_EXTRAS": os.environ.get("AXOLOTL_EXTRAS", ""), "AXOLOTL_EXTRAS": os.environ.get("AXOLOTL_EXTRAS", ""),
@@ -27,8 +28,11 @@ df_args = {
"CUDA": os.environ.get("CUDA", "126"), "CUDA": os.environ.get("CUDA", "126"),
"GITHUB_REF": os.environ.get("GITHUB_REF", "refs/heads/main"), "GITHUB_REF": os.environ.get("GITHUB_REF", "refs/heads/main"),
"GITHUB_SHA": os.environ.get("GITHUB_SHA", ""), "GITHUB_SHA": os.environ.get("GITHUB_SHA", ""),
"NIGHTLY_BUILD": os.environ.get("NIGHTLY_BUILD", ""),
"CODECOV_TOKEN": os.environ.get("CODECOV_TOKEN", ""), "CODECOV_TOKEN": os.environ.get("CODECOV_TOKEN", ""),
"HF_HOME": "/workspace/data/huggingface-cache/hub", "HF_HOME": "/workspace/data/huggingface-cache/hub",
"PYTHONUNBUFFERED": os.environ.get("PYTHONUNBUFFERED", "1"),
"DEEPSPEED_LOG_LEVEL": os.environ.get("DEEPSPEED_LOG_LEVEL", "WARNING"),
} }
dockerfile_contents = df_template.render(**df_args) dockerfile_contents = df_template.render(**df_args)

2
docker/Dockerfile-base
View File

@@ -43,7 +43,7 @@ ENV PATH="/root/miniconda3/envs/py${PYTHON_VERSION}/bin:${PATH}"
WORKDIR /workspace WORKDIR /workspace
RUN python3 -m pip install --upgrade pip && pip3 install -U packaging==23.2 setuptools==75.8.0 wheel psutil && \ RUN python3 -m pip install --upgrade pip && pip3 install -U packaging==26.0 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$CUDA && \ python3 -m pip install --no-cache-dir -U torch==${PYTORCH_VERSION}+cu${CUDA} torchvision --extra-index-url https://download.pytorch.org/whl/cu$CUDA && \
python3 -m pip cache purge python3 -m pip cache purge

2
docker/Dockerfile-base-nightly
View File

@@ -30,7 +30,7 @@ ENV PATH="/root/miniconda3/envs/py${PYTHON_VERSION}/bin:${PATH}"
WORKDIR /workspace 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==26.0 setuptools==75.8.0 wheel && \
python3 -m pip install --no-cache-dir -U torch --extra-index-url https://download.pytorch.org/whl/nightly/cu$CUDA && \ python3 -m pip install --no-cache-dir -U torch --extra-index-url https://download.pytorch.org/whl/nightly/cu$CUDA && \
python3 -m pip install --no-cache-dir "causal_conv1d @ git+https://github.com/Dao-AILab/causal-conv1d.git@main" && \ python3 -m pip install --no-cache-dir "causal_conv1d @ git+https://github.com/Dao-AILab/causal-conv1d.git@main" && \
python3 -m pip install --no-cache-dir "mamba_ssm @ git+https://github.com/state-spaces/mamba.git@main" && \ python3 -m pip install --no-cache-dir "mamba_ssm @ git+https://github.com/state-spaces/mamba.git@main" && \

2
examples/apertus/README.md
View File

@@ -15,7 +15,7 @@ This guide shows how to fine-tune it with Axolotl with multi-turn conversations
git clone https://github.com/axolotl-ai-cloud/axolotl.git git clone https://github.com/axolotl-ai-cloud/axolotl.git
cd axolotl cd axolotl
pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja pip3 install packaging==26.0 setuptools==75.8.0 wheel ninja
pip3 install --no-build-isolation -e '.[flash-attn]' pip3 install --no-build-isolation -e '.[flash-attn]'
# Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy # Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy

2
examples/arcee/README.md
View File

@@ -17,7 +17,7 @@ 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 git clone https://github.com/axolotl-ai-cloud/axolotl.git
cd axolotl cd axolotl
pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja pip3 install packaging==26.0 setuptools==75.8.0 wheel ninja
pip3 install --no-build-isolation -e '.[flash-attn]' pip3 install --no-build-isolation -e '.[flash-attn]'
# Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy # Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy

2
examples/devstral/README.md
View File

@@ -16,7 +16,7 @@ Thanks to the team at MistralAI for giving us early access to prepare for this r
```bash ```bash
# Ensure you have Pytorch installed (Pytorch 2.6.0 min) # Ensure you have Pytorch installed (Pytorch 2.6.0 min)
pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja pip3 install packaging==26.0 setuptools==75.8.0 wheel ninja
pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0' pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
``` ```

2
examples/gemma3n/README.md
View File

@@ -10,7 +10,7 @@ Gemma-3n is a family of multimodal models from Google found on [HuggingFace](htt
```bash ```bash
# Ensure you have Pytorch installed (Pytorch 2.6.0 min) # Ensure you have Pytorch installed (Pytorch 2.6.0 min)
pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja pip3 install packaging==26.0 setuptools==75.8.0 wheel ninja
pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0' pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
``` ```

2
examples/gpt-oss/README.md
View File

@@ -14,7 +14,7 @@ This guide shows how to fine-tune it with Axolotl with multi-turn conversations
```bash ```bash
# Ensure you have Pytorch installed (Pytorch 2.6.0 min) # Ensure you have Pytorch installed (Pytorch 2.6.0 min)
pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja pip3 install packaging==26.0 setuptools==75.8.0 wheel ninja
pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0' pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
``` ```

2
examples/granite4/README.md
View File

@@ -15,7 +15,7 @@ This guide shows how to fine-tune it with Axolotl with multi-turn conversations
git clone https://github.com/axolotl-ai-cloud/axolotl.git git clone https://github.com/axolotl-ai-cloud/axolotl.git
cd axolotl cd axolotl
pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja pip3 install packaging==26.0 setuptools==75.8.0 wheel ninja
pip3 install --no-build-isolation -e '.[flash-attn]' pip3 install --no-build-isolation -e '.[flash-attn]'
# Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy # Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy

2
examples/hunyuan/README.md
View File

@@ -13,7 +13,7 @@ Tencent released a family of opensource models called HunYuan with varying param
git clone https://github.com/axolotl-ai-cloud/axolotl.git git clone https://github.com/axolotl-ai-cloud/axolotl.git
cd axolotl cd axolotl
pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja pip3 install packaging==26.0 setuptools==75.8.0 wheel ninja
pip3 install --no-build-isolation -e '.[flash-attn]' pip3 install --no-build-isolation -e '.[flash-attn]'
# Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy # Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy

2
examples/magistral/README.md
View File

@@ -14,7 +14,7 @@ Thanks to the team at MistralAI for giving us early access to prepare for these
```bash ```bash
# Ensure you have Pytorch installed (Pytorch 2.7.0 min) # Ensure you have Pytorch installed (Pytorch 2.7.0 min)
pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja pip3 install packaging==26.0 setuptools==75.8.0 wheel ninja
pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0' pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
``` ```

2
examples/qwen3-next/README.md
View File

@@ -15,7 +15,7 @@ This guide shows how to fine-tune it with Axolotl with multi-turn conversations
git clone https://github.com/axolotl-ai-cloud/axolotl.git git clone https://github.com/axolotl-ai-cloud/axolotl.git
cd axolotl cd axolotl
pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja pip3 install packaging==26.0 setuptools==75.8.0 wheel ninja
pip3 install --no-build-isolation -e '.[flash-attn]' pip3 install --no-build-isolation -e '.[flash-attn]'
# Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy # Install CCE https://docs.axolotl.ai/docs/custom_integrations.html#cut-cross-entropy

2
examples/voxtral/README.md
View File

@@ -12,7 +12,7 @@ Thanks to the team at MistralAI for giving us early access to prepare for this r
```bash ```bash
# Ensure you have Pytorch installed (Pytorch 2.6.0 min) # Ensure you have Pytorch installed (Pytorch 2.6.0 min)
pip3 install packaging==23.2 setuptools==75.8.0 wheel ninja pip3 install packaging==26.0 setuptools==75.8.0 wheel ninja
pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0' pip3 install --no-build-isolation 'axolotl[flash-attn]>=0.12.0'
``` ```

2
pyproject.toml
View File

@@ -1,5 +1,5 @@
[build-system] [build-system]
requires = ["setuptools>=64", "wheel", "setuptools_scm>=8", "packaging==23.2"] requires = ["setuptools>=64", "wheel", "setuptools_scm>=8", "packaging==26.0"]
build-backend = "setuptools.build_meta" build-backend = "setuptools.build_meta"
[project] [project]

4
requirements.txt
View File

@@ -8,7 +8,7 @@ xformers>=0.0.23.post1
liger-kernel==0.6.4 liger-kernel==0.6.4
# END section # END section
packaging==23.2 packaging==26.0
huggingface_hub>=0.36.0 huggingface_hub>=0.36.0
peft>=0.18.1 peft>=0.18.1
@@ -72,4 +72,4 @@ axolotl-contribs-mit==0.0.6
# telemetry # telemetry
posthog==6.7.11 posthog==6.7.11
mistral-common==1.8.6 mistral-common==1.8.8

5
setup.py
View File

@@ -78,6 +78,11 @@ def parse_requirements(extras_require_map):
extras_require_map["vllm"] = ["vllm==0.11.1"] extras_require_map["vllm"] = ["vllm==0.11.1"]
if not install_xformers: if not install_xformers:
_install_requires.pop(_install_requires.index(xformers_version)) _install_requires.pop(_install_requires.index(xformers_version))
extras_require_map["vllm"] = ["vllm==0.13.0"]
if patch == 0:
extras_require_map["vllm"] = ["vllm==0.13.0"]
else:
extras_require_map["vllm"] = ["vllm==0.14.0"]
elif (major, minor) >= (2, 8): elif (major, minor) >= (2, 8):
extras_require_map.pop("fbgemm-gpu") extras_require_map.pop("fbgemm-gpu")
extras_require_map["fbgemm-gpu"] = ["fbgemm-gpu-genai==1.3.0"] extras_require_map["fbgemm-gpu"] = ["fbgemm-gpu-genai==1.3.0"]