[diffusion] CI: use dedicated HF token for accessing restricted models (#20620)

This commit is contained in:
Mick
2026-03-15 19:15:28 +08:00
committed by GitHub
parent 4f91b3069e
commit f07529b947
6 changed files with 15 additions and 0 deletions

View File

@@ -40,6 +40,8 @@ jobs:
run: bash scripts/ci/cuda/ci_install_dependency.sh diffusion
- name: Generate outputs
env:
HF_TOKEN: ${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }}
run: |
cd python
python -m sglang.multimodal_gen.test.scripts.gen_diffusion_ci_outputs \
@@ -74,6 +76,8 @@ jobs:
run: bash scripts/ci/cuda/ci_install_dependency.sh diffusion
- name: Generate outputs
env:
HF_TOKEN: ${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }}
run: |
cd python
python -m sglang.multimodal_gen.test.scripts.gen_diffusion_ci_outputs \

View File

@@ -422,6 +422,7 @@ jobs:
SGLANG_DIFFUSION_SLACK_TOKEN: ${{ secrets.SGLANG_DIFFUSION_SLACK_TOKEN }}
GITHUB_RUN_ID: ${{ github.run_id }}
GPU_CONFIG: "1-gpu-runner"
HF_TOKEN: ${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }}
timeout-minutes: 60
run: |
@@ -478,6 +479,7 @@ jobs:
SGLANG_DIFFUSION_SLACK_TOKEN: ${{ secrets.SGLANG_DIFFUSION_SLACK_TOKEN }}
GITHUB_RUN_ID: ${{ github.run_id }}
GPU_CONFIG: "2-gpu-runner"
HF_TOKEN: ${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }}
timeout-minutes: 60
run: |

View File

@@ -636,6 +636,7 @@ jobs:
-e MIOPEN_USER_DB_PATH=/sgl-data/miopen-cache \
-e HF_HUB_ENABLE_HF_TRANSFER=1 \
-e HF_HUB_DISABLE_SYMLINKS_WARNING=1 \
-e HF_TOKEN=${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }} \
-w /sglang-checkout/python \
ci_sglang python3 sglang/multimodal_gen/test/run_suite.py \
--suite 1-gpu \
@@ -765,6 +766,7 @@ jobs:
-e MIOPEN_USER_DB_PATH=/sgl-data/miopen-cache \
-e HF_HUB_ENABLE_HF_TRANSFER=1 \
-e HF_HUB_DISABLE_SYMLINKS_WARNING=1 \
-e HF_TOKEN=${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }} \
-w /sglang-checkout/python \
ci_sglang python3 sglang/multimodal_gen/test/run_suite.py \
--suite 2-gpu \

View File

@@ -640,6 +640,7 @@ jobs:
-e MIOPEN_USER_DB_PATH=/sgl-data/miopen-cache \
-e HF_HUB_ENABLE_HF_TRANSFER=1 \
-e HF_HUB_DISABLE_SYMLINKS_WARNING=1 \
-e HF_TOKEN=${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }} \
-w /sglang-checkout/python \
ci_sglang python3 sglang/multimodal_gen/test/run_suite.py \
--suite 1-gpu \
@@ -767,6 +768,7 @@ jobs:
-e MIOPEN_USER_DB_PATH=/sgl-data/miopen-cache \
-e HF_HUB_ENABLE_HF_TRANSFER=1 \
-e HF_HUB_DISABLE_SYMLINKS_WARNING=1 \
-e HF_TOKEN=${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }} \
-w /sglang-checkout/python \
ci_sglang python3 sglang/multimodal_gen/test/run_suite.py \
--suite 2-gpu \

View File

@@ -302,6 +302,7 @@ jobs:
env:
SGLANG_USE_MODELSCOPE: true
SGLANG_IS_IN_CI: true
HF_TOKEN: ${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }}
HF_ENDPOINT: https://hf-mirror.com
TORCH_EXTENSIONS_DIR: /tmp/torch_extensions
PYTORCH_NPU_ALLOC_CONF: "expandable_segments:True"
@@ -344,6 +345,7 @@ jobs:
env:
SGLANG_USE_MODELSCOPE: true
SGLANG_IS_IN_CI: true
HF_TOKEN: ${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }}
HF_ENDPOINT: https://hf-mirror.com
TORCH_EXTENSIONS_DIR: /tmp/torch_extensions
PYTORCH_NPU_ALLOC_CONF: "expandable_segments:True"
@@ -383,6 +385,7 @@ jobs:
env:
SGLANG_USE_MODELSCOPE: true
SGLANG_IS_IN_CI: true
HF_TOKEN: ${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }}
HF_ENDPOINT: https://hf-mirror.com
TORCH_EXTENSIONS_DIR: /tmp/torch_extensions
PYTORCH_NPU_ALLOC_CONF: "expandable_segments:True"

View File

@@ -1272,6 +1272,7 @@ jobs:
timeout-minutes: 240
env:
RUNAI_STREAMER_MEMORY_LIMIT: 0
HF_TOKEN: ${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }}
run: |
cd python
CONTINUE_ON_ERROR_FLAG=""
@@ -1330,6 +1331,7 @@ jobs:
timeout-minutes: 240
env:
RUNAI_STREAMER_MEMORY_LIMIT: 0
HF_TOKEN: ${{ secrets.SGLANG_DIFFUSION_CI_HF_TOKEN }}
run: |
cd python
CONTINUE_ON_ERROR_FLAG=""