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

deprecate torch 2.7.1 (#3339)

Browse Source
This commit is contained in:
Wing Lian
2026-01-01 06:52:45 -05:00
committed by GitHub
parent 2b199f9915
commit afe18ace35
11 changed files with 65 additions and 144 deletions
Download Patch File Download Diff File Expand all files Collapse all files

49
.github/workflows/base.yml vendored
View File

@@ -25,27 +25,6 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
include: include:
- cuda: "126"
cuda_version: 12.6.3
cudnn_version: ""
python_version: "3.11"
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"
cuda_version: 12.6.3
cudnn_version: ""
python_version: "3.11"
pytorch: 2.7.1
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: 2.7.1
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: "128"
cuda_version: 12.8.1 cuda_version: 12.8.1
cudnn_version: "" cudnn_version: ""
@@ -53,6 +32,13 @@ jobs:
pytorch: 2.8.0 pytorch: 2.8.0
torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX" torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
dockerfile: "Dockerfile-base" 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: "128" - cuda: "128"
cuda_version: 12.8.1 cuda_version: 12.8.1
cudnn_version: "" cudnn_version: ""
@@ -121,20 +107,6 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
include: include:
- cuda: "126"
cuda_version: 12.6.3
cudnn_version: ""
python_version: "3.11"
pytorch: 2.7.1
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: 2.7.1
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: "128"
cuda_version: 12.8.1 cuda_version: 12.8.1
cudnn_version: "" cudnn_version: ""
@@ -149,6 +121,13 @@ jobs:
pytorch: 2.9.1 pytorch: 2.9.1
torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX" torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
dockerfile: "Dockerfile-uv-base" 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: "130"
cuda_version: 13.0.0 cuda_version: 13.0.0
cudnn_version: "" cudnn_version: ""

64
.github/workflows/main.yml vendored
View File

@@ -15,21 +15,6 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
include: include:
- cuda: 126
cuda_version: 12.6.3
python_version: "3.11"
pytorch: 2.7.0
axolotl_extras:
- cuda: 126
cuda_version: 12.6.3
python_version: "3.11"
pytorch: 2.7.1
axolotl_extras: vllm
- cuda: 128
cuda_version: 12.8.1
python_version: "3.11"
pytorch: 2.7.1
axolotl_extras:
- cuda: 128 - cuda: 128
cuda_version: 12.8.1 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
@@ -46,6 +31,11 @@ jobs:
python_version: "3.11" python_version: "3.11"
pytorch: 2.9.1 pytorch: 2.9.1
axolotl_extras: axolotl_extras:
- cuda: 130
cuda_version: 13.0.0
python_version: "3.11"
pytorch: 2.9.1
axolotl_extras:
runs-on: axolotl-gpu-runner runs-on: axolotl-gpu-runner
steps: steps:
- name: Checkout - name: Checkout
@@ -92,27 +82,6 @@ jobs:
strategy: strategy:
matrix: matrix:
include: include:
- cuda: 126
cuda_version: 12.6.3
python_version: "3.11"
pytorch: 2.7.0
axolotl_extras:
- cuda: 126
cuda_version: 12.6.3
python_version: "3.11"
pytorch: 2.7.1
axolotl_extras:
is_latest:
- cuda: 126
cuda_version: 12.6.3
python_version: "3.11"
pytorch: 2.7.1
axolotl_extras: vllm
- cuda: 128
cuda_version: 12.8.1
python_version: "3.11"
pytorch: 2.7.1
axolotl_extras:
- cuda: 128 - cuda: 128
cuda_version: 12.8.1 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
@@ -129,6 +98,11 @@ jobs:
python_version: "3.11" python_version: "3.11"
pytorch: 2.9.1 pytorch: 2.9.1
axolotl_extras: axolotl_extras:
- cuda: 130
cuda_version: 13.0.0
python_version: "3.11"
pytorch: 2.9.1
axolotl_extras:
runs-on: axolotl-gpu-runner runs-on: axolotl-gpu-runner
steps: steps:
- name: Checkout - name: Checkout
@@ -170,24 +144,18 @@ jobs:
strategy: strategy:
matrix: matrix:
include: include:
- cuda: 126
cuda_version: 12.6.3
python_version: "3.11"
pytorch: 2.7.1
axolotl_extras:
is_latest:
- cuda: 126
cuda_version: 12.6.3
python_version: "3.11"
pytorch: 2.7.1
axolotl_extras: vllm
is_latest: true
- cuda: 128 - cuda: 128
cuda_version: 12.8.1 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
pytorch: 2.8.0 pytorch: 2.8.0
axolotl_extras: axolotl_extras:
is_latest: is_latest:
- cuda: 128
cuda_version: 12.8.1
python_version: "3.11"
pytorch: 2.9.1
axolotl_extras:
is_latest:
runs-on: axolotl-gpu-runner runs-on: axolotl-gpu-runner
steps: steps:
- name: Checkout - name: Checkout

9
.github/workflows/multi-gpu-e2e.yml vendored
View File

@@ -29,13 +29,6 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
include: include:
- cuda: 126
cuda_version: 12.6.3
python_version: "3.11"
pytorch: 2.7.1
axolotl_extras: vllm
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"
@@ -46,7 +39,7 @@ jobs:
- cuda: 128 - cuda: 128
cuda_version: 12.8.1 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
pytorch: 2.9.0 pytorch: 2.9.1
axolotl_extras: fbgemm-gpu axolotl_extras: fbgemm-gpu
num_gpus: 2 num_gpus: 2
nightly_build: "true" nightly_build: "true"

20
.github/workflows/nightlies.yml vendored
View File

@@ -12,16 +12,16 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
include: include:
- cuda: 126
cuda_version: 12.6.3
python_version: "3.11"
pytorch: 2.7.1
axolotl_extras:
- cuda: 128 - cuda: 128
cuda_version: 12.8.1 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
pytorch: 2.8.0 pytorch: 2.8.0
axolotl_extras: axolotl_extras:
- cuda: 128
cuda_version: 12.8.1
python_version: "3.11"
pytorch: 2.9.1
axolotl_extras:
runs-on: axolotl-gpu-runner runs-on: axolotl-gpu-runner
steps: steps:
- name: Checkout - name: Checkout
@@ -64,16 +64,16 @@ jobs:
strategy: strategy:
matrix: matrix:
include: include:
- cuda: 126
cuda_version: 12.6.3
python_version: "3.11"
pytorch: 2.7.1
axolotl_extras:
- cuda: 128 - cuda: 128
cuda_version: 12.8.1 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
pytorch: 2.8.0 pytorch: 2.8.0
axolotl_extras: axolotl_extras:
- cuda: 128
cuda_version: 12.8.1
python_version: "3.11"
pytorch: 2.9.1
axolotl_extras:
runs-on: axolotl-gpu-runner runs-on: axolotl-gpu-runner
steps: steps:
- name: Checkout - name: Checkout

16
.github/workflows/tests-nightly.yml vendored
View File

@@ -26,7 +26,7 @@ jobs:
max-parallel: 2 max-parallel: 2
matrix: matrix:
python_version: ["3.11"] python_version: ["3.11"]
pytorch_version: ["2.7.1", "2.8.0"] pytorch_version: ["2.8.0", "2.9.0", "2.9.1"]
timeout-minutes: 20 timeout-minutes: 20
steps: steps:
@@ -99,17 +99,17 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
include: include:
- cuda: 126 - cuda: 128
cuda_version: 12.6.3 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
pytorch: 2.7.1 pytorch: 2.8.0
num_gpus: 1 num_gpus: 1
axolotl_extras: axolotl_extras:
nightly_build: "true" 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: 2.8.0 pytorch: 2.9.1
num_gpus: 1 num_gpus: 1
axolotl_extras: axolotl_extras:
nightly_build: "true" nightly_build: "true"
@@ -148,10 +148,10 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
include: include:
- cuda: 126 - cuda: 128
cuda_version: 12.6.3 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
pytorch: 2.7.1 pytorch: 2.9.1
num_gpus: 2 num_gpus: 2
axolotl_extras: axolotl_extras:
nightly_build: "true" nightly_build: "true"

24
.github/workflows/tests.yml vendored
View File

@@ -55,7 +55,7 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
python_version: ["3.11"] python_version: ["3.11"]
pytorch_version: ["2.7.1", "2.8.0", "2.9.0"] pytorch_version: ["2.8.0", "2.9.0", "2.9.1"]
timeout-minutes: 20 timeout-minutes: 20
steps: steps:
@@ -145,7 +145,7 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
python_version: ["3.11"] python_version: ["3.11"]
pytorch_version: ["2.7.1", "2.8.0", "2.9.0"] pytorch_version: ["2.8.0", "2.9.0", "2.9.1"]
timeout-minutes: 20 timeout-minutes: 20
steps: steps:
@@ -303,18 +303,6 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
include: include:
- cuda: 126
cuda_version: 12.6.3
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: 128
cuda_version: 12.8.1 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
@@ -325,7 +313,7 @@ jobs:
- cuda: 128 - cuda: 128
cuda_version: 12.8.1 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
pytorch: 2.9.0 pytorch: 2.9.1
num_gpus: 1 num_gpus: 1
axolotl_extras: axolotl_extras:
steps: steps:
@@ -365,10 +353,10 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
include: include:
- cuda: 126 - cuda: 128
cuda_version: 12.6.3 cuda_version: 12.8.1
python_version: "3.11" python_version: "3.11"
pytorch: 2.7.1 pytorch: 2.9.1
num_gpus: 1 num_gpus: 1
axolotl_extras: axolotl_extras:
steps: steps:

2
README.md
View File

@@ -77,7 +77,7 @@ Features:
- NVIDIA GPU (Ampere or newer for `bf16` and Flash Attention) or AMD GPU - NVIDIA GPU (Ampere or newer for `bf16` and Flash Attention) or AMD GPU
- Python 3.11 - Python 3.11
- PyTorch ≥2.7.1 - PyTorch ≥2.8.0
### Google Colab ### Google Colab

16
docs/docker.qmd
View File

@@ -32,11 +32,8 @@ main-base-py{python_version}-cu{cuda_version}-{pytorch_version}
Tags examples: Tags examples:
- `main-base-py3.11-cu128-2.7.1` - `main-base-py3.11-cu128-2.8.0`
- `main-base-py3.11-cu126-2.7.1` - `main-base-py3.11-cu128-2.9.1`
- `main-base-py3.11-cu126-2.7.0`
- `main-base-py3.11-cu126-2.6.0`
- `main-base-py3.11-cu124-2.6.0`
## Main ## Main
@@ -74,15 +71,12 @@ There may be some extra tags appended to the image, like `-vllm` which installs
Tags examples: Tags examples:
- `main-py3.11-cu128-2.7.1` - `main-py3.11-cu128-2.8.0`
- `main-py3.11-cu126-2.7.1` - `main-py3.11-cu128-2.9.1`
- `main-py3.11-cu126-2.7.0`
- `main-py3.11-cu126-2.6.0`
- `main-py3.11-cu124-2.6.0`
- `main-latest` - `main-latest`
- `main-20250303-py3.11-cu124-2.6.0` - `main-20250303-py3.11-cu124-2.6.0`
- `main-20250303-py3.11-cu126-2.6.0` - `main-20250303-py3.11-cu126-2.6.0`
- `0.10.1` - `0.12.0`
## Cloud ## Cloud

4
docs/installation.qmd
View File

@@ -26,7 +26,7 @@ Follow the instructions at: [https://pytorch.org/get-started/locally/](https://p
::: :::
::: {.callout-important} ::: {.callout-important}
For Blackwell GPUs, please use Pytorch 2.7.0 and CUDA 12.8. For Blackwell GPUs, please use Pytorch 2.9.1 and CUDA 12.8.
::: :::
### PyPI Installation (Recommended) {#sec-pypi} ### PyPI Installation (Recommended) {#sec-pypi}
@@ -111,7 +111,7 @@ docker run --privileged --gpus '"all"' --shm-size 10g --rm -it \
::: :::
::: {.callout-important} ::: {.callout-important}
For Blackwell GPUs, please use `axolotlai/axolotl:main-py3.11-cu128-2.7.0` or the cloud variant `axolotlai/axolotl-cloud:main-py3.11-cu128-2.7.0`. For Blackwell GPUs, please use `axolotlai/axolotl:main-py3.11-cu128-2.9.1` or the cloud variant `axolotlai/axolotl-cloud:main-py3.11-cu128-2.9.1`.
::: :::
Please refer to the [Docker documentation](docker.qmd) for more information on the different Docker images that are available. Please refer to the [Docker documentation](docker.qmd) for more information on the different Docker images that are available.

3
src/axolotl/cli/cloud/baseten/template/train_sft.py
View File

@@ -24,8 +24,7 @@ if launcher_args:
launcher_args_str = "-- " + " ".join(launcher_args) launcher_args_str = "-- " + " ".join(launcher_args)
# 1. Define a base image for your training job # 1. Define a base image for your training job
# must use torch 2.7.0 for vllm BASE_IMAGE = "axolotlai/axolotl:main-py3.11-cu128-2.9.1"
BASE_IMAGE = "axolotlai/axolotl:main-py3.11-cu126-2.7.1"
# 2. Define the Runtime Environment for the Training Job # 2. Define the Runtime Environment for the Training Job
# This includes start commands and environment variables.a # This includes start commands and environment variables.a

2
src/axolotl/cli/cloud/modal_.py
View File

@@ -82,7 +82,7 @@ class ModalCloud(Cloud):
return res return res
def get_image(self): def get_image(self):
docker_tag = "main-py3.11-cu126-2.7.1" docker_tag = "main-py3.11-cu128-2.9.1"
if self.config.docker_tag: if self.config.docker_tag:
docker_tag = self.config.docker_tag docker_tag = self.config.docker_tag
docker_image = f"axolotlai/axolotl:{docker_tag}" docker_image = f"axolotlai/axolotl:{docker_tag}"