biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[docs] Add docs for new RL flows (#36188)

Browse Source 
Signed-off-by: ahao-anyscale <ahao@anyscale.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Aaron Hao
2026-03-18 02:04:26 -07:00
committed by GitHub
parent fad09e8a1f
commit 47a1f11bff
18 changed files with 514 additions and 760 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
.buildkite/test_areas/distributed.yaml
View File

@@ -82,7 +82,7 @@ steps:
- label: Distributed Torchrun + Examples (4 GPUs)
timeout_in_minutes: 30
working_dir: "/vllm-workspace/tests"
working_dir: "/vllm-workspace"
num_devices: 4
source_file_dependencies:
- vllm/distributed/
@@ -90,33 +90,28 @@ steps:
- tests/distributed/test_torchrun_example_moe.py
- examples/offline_inference/rlhf.py
- examples/offline_inference/rlhf_colocate.py
- examples/offline_inference/new_weight_syncing/
- examples/rl/
- tests/examples/offline_inference/data_parallel.py
commands:
# https://github.com/NVIDIA/nccl/issues/1838
- export NCCL_CUMEM_HOST_ENABLE=0
# test with torchrun tp=2 and external_dp=2
- torchrun --nproc-per-node=4 distributed/test_torchrun_example.py
- torchrun --nproc-per-node=4 tests/distributed/test_torchrun_example.py
# test with torchrun tp=2 and pp=2
- PP_SIZE=2 torchrun --nproc-per-node=4 distributed/test_torchrun_example.py
- PP_SIZE=2 torchrun --nproc-per-node=4 tests/distributed/test_torchrun_example.py
# test with torchrun tp=4 and dp=1
- TP_SIZE=4 torchrun --nproc-per-node=4 distributed/test_torchrun_example_moe.py
- TP_SIZE=4 torchrun --nproc-per-node=4 tests/distributed/test_torchrun_example_moe.py
# test with torchrun tp=2, pp=2 and dp=1
- PP_SIZE=2 TP_SIZE=2 torchrun --nproc-per-node=4 distributed/test_torchrun_example_moe.py
- PP_SIZE=2 TP_SIZE=2 torchrun --nproc-per-node=4 tests/distributed/test_torchrun_example_moe.py
# test with torchrun tp=1 and dp=4 with ep
- DP_SIZE=4 ENABLE_EP=1 torchrun --nproc-per-node=4 distributed/test_torchrun_example_moe.py
- DP_SIZE=4 ENABLE_EP=1 torchrun --nproc-per-node=4 tests/distributed/test_torchrun_example_moe.py
# test with torchrun tp=2 and dp=2 with ep
- TP_SIZE=2 DP_SIZE=2 ENABLE_EP=1 torchrun --nproc-per-node=4 distributed/test_torchrun_example_moe.py
- TP_SIZE=2 DP_SIZE=2 ENABLE_EP=1 torchrun --nproc-per-node=4 tests/distributed/test_torchrun_example_moe.py
# test with internal dp
- python3 ../examples/offline_inference/data_parallel.py --enforce-eager
# OLD rlhf examples
- cd ../examples/offline_inference
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 python3 rlhf.py
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 RAY_DEDUP_LOGS=0 python3 rlhf_colocate.py
# NEW rlhf examples
- cd new_weight_syncing
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 python3 rlhf_nccl.py
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 python3 rlhf_ipc.py
- python3 examples/offline_inference/data_parallel.py --enforce-eager
# rlhf examples
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 python3 examples/rl/rlhf_nccl.py
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 python3 examples/rl/rlhf_ipc.py
- label: Distributed DP Tests (4 GPUs)
timeout_in_minutes: 30