Merge branch 'volcengine:main' into main

Author: sunyi0505

.github/PULL_REQUEST_TEMPLATE.md

@@ -37,4 +37,4 @@
 - [ ] Apply [pre-commit checks](https://github.com/volcengine/verl/blob/main/CONTRIBUTING.md#code-linting-and-formatting): `pre-commit install && pre-commit run --all-files --show-diff-on-failure --color=always`
 - [ ] Add / Update [the documentation](https://github.com/volcengine/verl/tree/main/docs).
 - [ ] Add unit or end-to-end test(s) to [the CI workflow](https://github.com/volcengine/verl/tree/main/.github/workflows) to cover all the code. If not feasible, explain why: ...
-- [ ] Once your PR is ready for CI, send a message in [the `ci-request` channel](https://verl-project.slack.com/archives/C091TCESWB1) in [the `verl` Slack workspace](https://join.slack.com/t/verl-project/shared_invite/zt-3855yhg8g-CTkqXu~hKojPCmo7k_yXTQ).
+- [ ] Once your PR is ready for CI, send a message in [the `ci-request` channel](https://verl-project.slack.com/archives/C091TCESWB1) in [the `verl` Slack workspace](https://join.slack.com/t/verl-project/shared_invite/zt-3855yhg8g-CTkqXu~hKojPCmo7k_yXTQ). (If not accessible, please try [the Feishu group (飞书群)](https://applink.larkoffice.com/client/chat/chatter/add_by_link?link_token=772jd4f1-cd91-441e-a820-498c6614126a).)

.github/workflows/checkpoint_converter.yml

@@ -12,7 +12,7 @@
 # - `special_sanity`: a suite of quick sanity tests
 # - `special_standalone`: a set of test that are designed to run in dedicated environments
 
-# Accelerators for tests 
+# Accelerators for tests
 # - By default tests are run with GPU available, except for the ones under `special_npu`, and any test script whose name ends with `on_cpu.py`.
 # - For test scripts with `on_cpu.py` name suffix would be tested on CPU resources in linux environment.
 
@@ -29,8 +29,6 @@
 #     - new workflow yaml is added to `.github/workflows`
 #     - new tests are added to workflow mentioned in 2.
 
-
-
 name: checkpoint_converter
 # latest version: Megatron-LM core_r0.11.0 https://github.com/NVIDIA/Megatron-LM/tree/core_r0.11.0
 
@@ -64,14 +62,13 @@ on:
       - "verl/trainer/main_ppo.py"
       - "verl/trainer/config/ppo_megatron_trainer.yaml"
 
-
 # Cancel jobs on the same ref if a new one is triggered
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}
   cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
 
 # Declare permissions just read content.
-permissions: 
+permissions:
   contents: read
 
 jobs:
@@ -84,15 +81,20 @@ jobs:
       NO_PROXY: "localhost,127.0.0.1"
       HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
     container:
-      image: verlai/verl:app-verl0.4-sglang0.4.6.post5-vllm0.8.5-mcore0.12.2-te2.2
+      image: verlai/verl:app-verl0.5-sglang0.4.8-mcore0.12.2-te2.2
       options: --gpus all --shm-size=10g
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
         with:
-            fetch-depth: 0
+          fetch-depth: 0
       - name: Install the current repository
         run: |
           pip3 install -e .[test]
+      - name: Download Model to Use
+        run: |
+          huggingface-cli download Qwen/Qwen2.5-0.5B --local-dir ${HOME}/models/Qwen/Qwen2.5-0.5B
+          huggingface-cli download deepseek-ai/deepseek-coder-1.3b-instruct --local-dir ${HOME}/models/deepseek-ai/deepseek-coder-1.3b-instruct
+          export HF_HUB_OFFLINE=1
       - name: Running Huggingface to Megatron dist_ckpt converter (Qwen/Qwen2.5-0.5B)
         run: |
           ray stop --force
@@ -114,12 +116,12 @@ jobs:
       HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
       HF_ENDPOINT: "https://hf-mirror.com"
     container:
-      image: verlai/verl:app-verl0.4-sglang0.4.6.post5-vllm0.8.5-mcore0.12.2-te2.2
+      image: verlai/verl:app-verl0.5-sglang0.4.8-mcore0.12.2-te2.2
       options: --gpus all --shm-size=10g
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
         with:
-            fetch-depth: 0
+          fetch-depth: 0
       - name: Install the current repository
         run: |
           pip3 install -e .[test]

.github/workflows/cpu_unit_tests.yml

@@ -59,17 +59,20 @@ permissions:
 
 jobs:
   cpu_unit_tests:
-    runs-on: ubuntu-latest
-    timeout-minutes: 10 # Increase this timeout value as needed
-    strategy:
-      matrix:
-        python-version: ["3.10"]
+    runs-on: [L20x8]
+    timeout-minutes: 20 # Increase this timeout value as needed
+    env:
+      HTTP_PROXY: ${{ secrets.PROXY_HTTP }}
+      HTTPS_PROXY: ${{ secrets.PROXY_HTTPS }}
+      NO_PROXY: "localhost,127.0.0.1,hf-mirror.com"
+      HF_ENDPOINT: "https://hf-mirror.com"
+      HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
+    container:
+      image: verlai/verl:app-verl0.5-vllm0.9.1-mcore0.12.2-te2.2
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
-      - name: Set up Python ${{ matrix.python-version }}
-        uses: actions/setup-python@0b93645e9fea7318ecaed2b359559ac225c90a2b # v5.3.0
         with:
-          python-version: ${{ matrix.python-version }}
+          fetch-depth: 0
       - name: Install the current repository
         run: |
           pip install -e .[test,prime,geo]
@@ -80,4 +83,4 @@ jobs:
           python3 examples/data_preprocess/geo3k.py
           echo '[pytest]' > pytest.ini
           echo 'python_files = *_on_cpu.py' >> pytest.ini
-          pytest -s -x --asyncio-mode=auto tests/
+          pytest -s -x --asyncio-mode=auto tests/

.github/workflows/e2e_ascend.yml

@@ -72,7 +72,7 @@ jobs:
   test:
     name: verl Ascend test (self-host)
     runs-on: [self-hosted, npu-0]
-    timeout-minutes: 30 # Increase this timeout value as needed
+    timeout-minutes: 40 # Increase this timeout value as needed
     container:
       image: crispig/verl_npu:cann8.1rc1-py3.10-torch2.5.1-vllm-ascend0.7.3.post1-250616
       volumes:

.github/workflows/e2e_dapo.yml

@@ -94,7 +94,7 @@ jobs:
       HF_ENDPOINT: "https://hf-mirror.com"
       HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
     container:
-      image: verlai/verl:app-verl0.4-vllm0.8.5-mcore0.12.2-te2.2
+      image: verlai/verl:app-verl0.5-vllm0.9.1-mcore0.12.2-te2.2
       options: --gpus all --shm-size=10g
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2

.github/workflows/e2e_eval_aime24.yml

@@ -88,7 +88,7 @@ permissions:
   contents: read
 
 env:
-  IMAGE: "verl-ci-cn-beijing.cr.volces.com/verlai/verl:app-verl0.4-vllm0.8.5-mcore0.12.2-te2.2"
+  IMAGE: "verl-ci-cn-beijing.cr.volces.com/verlai/verl:app-verl0.5-vllm0.9.1-mcore0.12.2-te2.2"
   DYNAMIC_RUNNER_ENDPOINT: "https://sd10g3clalm04ug7alq90.apigateway-cn-beijing.volceapi.com/runner"
 
 jobs:

.github/workflows/e2e_ppo_trainer.yml

@@ -87,7 +87,7 @@ jobs:
       HF_ENDPOINT: "https://hf-mirror.com"
       HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
     container:
-      image: verlai/verl:app-verl0.4-vllm0.8.5-mcore0.12.2-te2.2
+      image: verlai/verl:app-verl0.5-vllm0.9.1-mcore0.12.2-te2.2
       options: --gpus all --shm-size=10g
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
@@ -211,6 +211,12 @@ jobs:
         run: |
           ray stop --force
           FUSED_KERNEL=True FUSED_KERNEL_BACKEND=triton bash tests/special_e2e/ppo_trainer/run_model_reward.sh
+      - name: Running GSM8K E2E training tests on vllm async
+        run: |
+          ray stop --force
+          export VLLM_USE_V1=1
+          ray start --head
+          TOTAL_TRAIN_STEPS=2 ENGINE=vllm ROLLOUT_MODE=async bash tests/special_e2e/ppo_trainer/run_function_reward.sh
 
   e2e_ppo_trainer_vllm_vlm:
     runs-on: [L20x8]
@@ -223,15 +229,15 @@ jobs:
       HF_ENDPOINT: "https://hf-mirror.com"
       HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
     container:
-      image: verlai/verl:app-verl0.4-vllm0.8.5-mcore0.12.2-te2.2
+      image: verlai/verl:app-verl0.5-vllm0.9.1-mcore0.12.2-te2.2
       options: --gpus all --shm-size=50g # Visual dataloader requires large memory
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
         with:
           fetch-depth: 0
       - name: Install the current repository
         run: |
-          pip3 install -e .[test,gpu,vllm,geo,trl]
+          pip3 install --no-deps -e .[test,gpu,vllm,geo,trl]
           pip install "transformers[hf_xet]<4.53.0" # Fix for transformers 4.53.0
       # Geo3k
       - name: Prepare GEO3K dataset
@@ -280,7 +286,7 @@ jobs:
       HF_ENDPOINT: "https://hf-mirror.com"
       HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
     container:
-      image: verlai/verl:app-verl0.4-sglang0.4.6.post5-vllm0.8.5-mcore0.12.2-te2.2
+      image: verlai/verl:app-verl0.5-sglang0.4.8-mcore0.12.2-te2.2
       options: --gpus all --shm-size=10g
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
@@ -301,12 +307,6 @@ jobs:
         run: |
           ray stop --force
           TOTAL_TRAIN_STEPS=2 ENGINE=sglang ROLLOUT_MODE=async bash tests/special_e2e/ppo_trainer/run_function_reward.sh
-      - name: Running GSM8K E2E training tests on vllm async
-        run: |
-          ray stop --force
-          export VLLM_USE_V1=1
-          ray start --head
-          TOTAL_TRAIN_STEPS=2 ENGINE=vllm ROLLOUT_MODE=async bash tests/special_e2e/ppo_trainer/run_function_reward.sh
 
   e2e_ppo_trainer_sglang_multiturn_with_tool:
     runs-on: [L20x8]
@@ -319,7 +319,7 @@ jobs:
       HF_ENDPOINT: "https://hf-mirror.com"
       HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
     container:
-      image: verlai/verl:app-verl0.4-sglang0.4.6.post5-vllm0.8.5-mcore0.12.2-te2.2
+      image: verlai/verl:app-verl0.5-sglang0.4.8-mcore0.12.2-te2.2
       options: --gpus all --shm-size=10g
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
@@ -352,7 +352,7 @@ jobs:
       HF_ENDPOINT: "https://hf-mirror.com"
       HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
     container:
-      image: verlai/verl:app-verl0.4-sglang0.4.6.post5-vllm0.8.5-mcore0.12.2-te2.2
+      image: verlai/verl:app-verl0.5-sglang0.4.8-mcore0.12.2-te2.2
       options: --gpus all --shm-size=50g # Visual dataloader requires large memory
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
@@ -409,7 +409,7 @@ jobs:
       HF_ENDPOINT: "https://hf-mirror.com"
       HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
     container:
-      image: verlai/verl:app-verl0.4-sglang0.4.6.post5-vllm0.8.5-mcore0.12.2-te2.2
+      image: verlai/verl:app-verl0.5-sglang0.4.8-mcore0.12.2-te2.2
       options: --gpus all --shm-size=10g
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2

.github/workflows/e2e_ppo_trainer_megatron_sglang.yml

@@ -12,7 +12,7 @@
 # - `special_sanity`: a suite of quick sanity tests
 # - `special_standalone`: a set of test that are designed to run in dedicated environments
 
-# Accelerators for tests 
+# Accelerators for tests
 # - By default tests are run with GPU available, except for the ones under `special_npu`, and any test script whose name ends with `on_cpu.py`.
 # - For test scripts with `on_cpu.py` name suffix would be tested on CPU resources in linux environment.
 
@@ -85,7 +85,7 @@ permissions:
   contents: read
 
 env:
-  IMAGE: "verl-ci-cn-beijing.cr.volces.com/verlai/verl:app-verl0.4-sglang0.4.6.post5-vllm0.8.5-mcore0.12.2-te2.2"
+  IMAGE: "verl-ci-cn-beijing.cr.volces.com/verlai/verl:app-verl0.5-sglang0.4.8-mcore0.12.2-te2.2"
   DYNAMIC_RUNNER_ENDPOINT: "https://sd10g3clalm04ug7alq90.apigateway-cn-beijing.volceapi.com/runner"
 
 jobs:
@@ -98,7 +98,7 @@ jobs:
     steps:
       - uses: actions/checkout@v4
       - id: create-runner
-        uses: volcengine/vemlp-github-runner@v1 
+        uses: volcengine/vemlp-github-runner@v1
         with:
           mode: "create"
           faas-url: "${{ env.DYNAMIC_RUNNER_ENDPOINT }}"
@@ -234,6 +234,10 @@ jobs:
       - name: Prepare GSM8K dataset
         run: |
           python3 examples/data_preprocess/gsm8k.py
+      - name: Download Model to Use
+        run: |
+          huggingface-cli download Qwen/Qwen2.5-0.5B --local-dir ${HOME}/models/Qwen/Qwen2.5-0.5B
+          export HF_HUB_OFFLINE=1
       - name: Prepare dist_ckpt of Qwen2.5-0.5B, uneven layer distribution only supports dist_ckpt
         run: |
           python3 scripts/converter_hf_to_mcore.py --hf_model_path ${HOME}/models/Qwen/Qwen2.5-0.5B --output_path checkpoints/verl-test/qwen2.5-0.5b-megatron
@@ -341,26 +345,29 @@ jobs:
       - name: Running Geo3k E2E training tests with 3D parallelism on 8 L20 GPUs with Megatron (Qwen)
         run: |
           ray stop --force
-          ENGINE=sglang TRAIN_FILES=${HOME}/data/geo3k/train.parquet VAL_FILES=${HOME}/data/geo3k/test.parquet MAX_PROMPT_LENGTH=1024 MAX_RESPONSE_LENGTH=2048  MODEL_ID=Qwen/Qwen2.5-VL-3B-Instruct ADV_ESTIMATOR=grpo USE_DYNAMIC_BSZ=False SKIP_SAVE_HF_MODEL=1 COMMON_PP=4 COMMON_VPP=null COMMON_CP=1 COMMON_TP=2 USE_DIST_CKPT=true DIST_CKPT_PATH=checkpoints/verl-test/qwen2.5-vl-3b-megatron bash tests/special_e2e/run_ppo_trainer_megatron.sh 
+          ENGINE=sglang TRAIN_FILES=${HOME}/data/geo3k/train.parquet VAL_FILES=${HOME}/data/geo3k/test.parquet MAX_PROMPT_LENGTH=1024 MAX_RESPONSE_LENGTH=2048  MODEL_ID=Qwen/Qwen2.5-VL-3B-Instruct ADV_ESTIMATOR=grpo USE_DYNAMIC_BSZ=False SKIP_SAVE_HF_MODEL=1 COMMON_PP=4 COMMON_VPP=null COMMON_CP=1 COMMON_TP=2 USE_DIST_CKPT=true DIST_CKPT_PATH=checkpoints/verl-test/qwen2.5-vl-3b-megatron bash tests/special_e2e/run_ppo_trainer_megatron.sh
       - name: clean up
         run: |
           rm -rf checkpoints
 
   cleanup:
     runs-on: ubuntu-latest
-    needs: [setup, 
-      e2e_ppo_trainer_megatron-deepseek, 
-      e2e_ppo_trainer_megatron-qwen3,
-      e2e_ppo_trainer_megatron-different-train-infer-tp-qwen-tie-embedding,
-      e2e_ppo_trainer_megatron-qwen-override-transformer-config,
-      e2e_ppo_trainer_megatron-deepseek-override-transformer-config,
-      e2e_ppo_trainer_megatron-moe-expert-parallel,
-      e2e_ppo_trainer_megatron-qwen2_5vl-3b]
+    needs:
+      [
+        setup,
+        e2e_ppo_trainer_megatron-deepseek,
+        e2e_ppo_trainer_megatron-qwen3,
+        e2e_ppo_trainer_megatron-different-train-infer-tp-qwen-tie-embedding,
+        e2e_ppo_trainer_megatron-qwen-override-transformer-config,
+        e2e_ppo_trainer_megatron-deepseek-override-transformer-config,
+        e2e_ppo_trainer_megatron-moe-expert-parallel,
+        e2e_ppo_trainer_megatron-qwen2_5vl-3b,
+      ]
     if: always()
     steps:
       - id: destroy-runner
         uses: volcengine/vemlp-github-runner@v1
         with:
           mode: "destroy"
           faas-url: "${{ env.DYNAMIC_RUNNER_ENDPOINT }}"
-          mlp-task-id: "${{ needs.setup.outputs.mlp-task-id }}"
+          mlp-task-id: "${{ needs.setup.outputs.mlp-task-id }}"

.github/workflows/e2e_ppo_trainer_megatron_vllm.yml

@@ -12,7 +12,7 @@
 # - `special_sanity`: a suite of quick sanity tests
 # - `special_standalone`: a set of test that are designed to run in dedicated environments
 
-# Accelerators for tests 
+# Accelerators for tests
 # - By default tests are run with GPU available, except for the ones under `special_npu`, and any test script whose name ends with `on_cpu.py`.
 # - For test scripts with `on_cpu.py` name suffix would be tested on CPU resources in linux environment.
 
@@ -85,7 +85,7 @@ permissions:
   contents: read
 
 env:
-  IMAGE: "verl-ci-cn-beijing.cr.volces.com/verlai/verl:app-verl0.4-vllm0.8.5-mcore0.12.2-te2.2"
+  IMAGE: "verl-ci-cn-beijing.cr.volces.com/verlai/verl:app-verl0.5-vllm0.9.1-mcore0.12.2-te2.2"
   DYNAMIC_RUNNER_ENDPOINT: "https://sd10g3clalm04ug7alq90.apigateway-cn-beijing.volceapi.com/runner"
 
 jobs:
@@ -98,7 +98,7 @@ jobs:
     steps:
       - uses: actions/checkout@v4
       - id: create-runner
-        uses: volcengine/vemlp-github-runner@v1 
+        uses: volcengine/vemlp-github-runner@v1
         with:
           mode: "create"
           faas-url: "${{ env.DYNAMIC_RUNNER_ENDPOINT }}"
@@ -214,7 +214,7 @@ jobs:
       - name: Running GSM8K E2E training tests with 3D parallelism on 8 L20 GPUs with Megatron (Qwen) with  train tp < infer tp
         run: |
           ray stop --force
-          VAL_BEFORE_TRAIN=True TEST_FREQ=1 SAVE_FREQ=1 TRAIN_TP=1 INFER_TP=2 MODEL_ID=Qwen/Qwen2.5-1.5B bash tests/special_e2e/run_ppo_trainer_megatron.sh
+          VAL_BEFORE_TRAIN=True TEST_FREQ=1 SAVE_FREQ=1 TRAIN_TP=1 INFER_TP=2 ALL_OFFLOAD=True MODEL_ID=Qwen/Qwen2.5-1.5B bash tests/special_e2e/run_ppo_trainer_megatron.sh
       - name: clean up
         run: |
           rm -rf checkpoints
@@ -238,6 +238,10 @@ jobs:
       - name: Prepare GSM8K dataset
         run: |
           python3 examples/data_preprocess/gsm8k.py
+      - name: Download Model to Use
+        run: |
+          huggingface-cli download Qwen/Qwen2.5-0.5B --local-dir ${HOME}/models/Qwen/Qwen2.5-0.5B
+          export HF_HUB_OFFLINE=1
       - name: Prepare dist_ckpt of Qwen2.5-0.5B, uneven layer distribution only supports dist_ckpt
         run: |
           python3 scripts/converter_hf_to_mcore.py --hf_model_path ${HOME}/models/Qwen/Qwen2.5-0.5B --output_path checkpoints/verl-test/qwen2.5-0.5b-megatron
@@ -347,26 +351,29 @@ jobs:
       - name: Running Geo3k E2E training tests with 3D parallelism on 8 L20 GPUs with Megatron (Qwen)
         run: |
           ray stop --force
-          TRAIN_FILES=${HOME}/data/geo3k/train.parquet VAL_FILES=${HOME}/data/geo3k/test.parquet MAX_PROMPT_LENGTH=1024 MAX_RESPONSE_LENGTH=2048  MODEL_ID=Qwen/Qwen2.5-VL-3B-Instruct ADV_ESTIMATOR=grpo USE_DYNAMIC_BSZ=False SKIP_SAVE_HF_MODEL=1 COMMON_PP=4 COMMON_VPP=null COMMON_CP=1 COMMON_TP=2 USE_DIST_CKPT=true DIST_CKPT_PATH=checkpoints/verl-test/qwen2.5-vl-3b-megatron bash tests/special_e2e/run_ppo_trainer_megatron.sh 
+          TRAIN_FILES=${HOME}/data/geo3k/train.parquet VAL_FILES=${HOME}/data/geo3k/test.parquet MAX_PROMPT_LENGTH=1024 MAX_RESPONSE_LENGTH=2048  MODEL_ID=Qwen/Qwen2.5-VL-3B-Instruct ADV_ESTIMATOR=grpo USE_DYNAMIC_BSZ=False SKIP_SAVE_HF_MODEL=1 COMMON_PP=4 COMMON_VPP=null COMMON_CP=1 COMMON_TP=2 USE_DIST_CKPT=true DIST_CKPT_PATH=checkpoints/verl-test/qwen2.5-vl-3b-megatron bash tests/special_e2e/run_ppo_trainer_megatron.sh
       - name: clean up
         run: |
           rm -rf checkpoints
 
   cleanup:
     runs-on: ubuntu-latest
-    needs: [setup, 
-      e2e_ppo_trainer_megatron-deepseek, 
-      e2e_ppo_trainer_megatron-qwen3,
-      e2e_ppo_trainer_megatron-different-train-infer-tp-qwen-tie-embedding,
-      e2e_ppo_trainer_megatron-qwen-override-transformer-config,
-      e2e_ppo_trainer_megatron-deepseek-override-transformer-config,
-      e2e_ppo_trainer_megatron-moe-expert-parallel,
-      e2e_ppo_trainer_megatron-qwen2_5vl-3b]
+    needs:
+      [
+        setup,
+        e2e_ppo_trainer_megatron-deepseek,
+        e2e_ppo_trainer_megatron-qwen3,
+        e2e_ppo_trainer_megatron-different-train-infer-tp-qwen-tie-embedding,
+        e2e_ppo_trainer_megatron-qwen-override-transformer-config,
+        e2e_ppo_trainer_megatron-deepseek-override-transformer-config,
+        e2e_ppo_trainer_megatron-moe-expert-parallel,
+        e2e_ppo_trainer_megatron-qwen2_5vl-3b,
+      ]
     if: always()
     steps:
       - id: destroy-runner
         uses: volcengine/vemlp-github-runner@v1
         with:
           mode: "destroy"
           faas-url: "${{ env.DYNAMIC_RUNNER_ENDPOINT }}"
-          mlp-task-id: "${{ needs.setup.outputs.mlp-task-id }}"
+          mlp-task-id: "${{ needs.setup.outputs.mlp-task-id }}"