Skip to content

Update B200 Dsv4 configs #1655

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Oseltamivir merged 8 commits into from
Jun 5, 2026
+10 −1
Merged

Update B200 Dsv4 configs #1655

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
8b3e665
Enable EPLB for DEP configs
wzhao18 Jun 3, 2026
bb31d48
Perf changelog
wzhao18 Jun 3, 2026
c361dad
Update to nightly image
wzhao18 Jun 3, 2026
94e89cc
Add NCCL_NET_PLUGIN and UCX_TLS exports
wzhao18 Jun 3, 2026
eba41e5
Merge branch 'main' into wzhao/dsv4-b200-eplb
wzhao18 Jun 3, 2026
31a12b0
Enhance EPLB_ARGS with eplb-config
wzhao18 Jun 3, 2026
783dfe8
Clean up environment variable exports in script
wzhao18 Jun 3, 2026
e1b0750
Merge branch 'main' into wzhao/dsv4-b200-eplb
Oseltamivir Jun 5, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .github/configs/nvidia-master.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1756,7 +1756,7 @@ dsv4-fp4-b200-sglang:
- { tp: 8, ep: 8, dp-attn: true, conc-start: 256, conc-end: 512 }

dsv4-fp4-b200-vllm:
image: vllm/vllm-openai:v0.22.0
image: vllm/vllm-openai:nightly-3f0a91bb96f8d72e0498b95c166e817deae14d62
model: deepseek-ai/DeepSeek-V4-Pro
model-prefix: dsv4
runner: b200-dsv4
Expand Down
3 changes: 3 additions & 0 deletions benchmarks/single_node/fixed_seq_len/dsv4_fp4_b200_vllm.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,8 +43,10 @@ fi

GMU_ARGS=()
MOE_ARGS=()
EPLB_ARGS=()
if [ "${DP_ATTENTION}" = "true" ]; then
MOE_ARGS=(--moe-backend deep_gemm_mega_moe)
EPLB_ARGS=(--enable-eplb --eplb-config '{"communicator":"torch_nccl", "use_async": false}')
fi

if [ "${ISL}" -eq 8192 ] && [ "${CONC}" -le 128 ]; then
Expand Down Expand Up @@ -78,6 +80,7 @@ vllm serve "$MODEL" --host 0.0.0.0 --port "$PORT" \
"${EP_ARGS[@]}" \
"${GMU_ARGS[@]}" \
"${MOE_ARGS[@]}" \
"${EPLB_ARGS[@]}" \
--compilation-config '{"cudagraph_mode":"FULL_AND_PIECEWISE","custom_ops":["all"]}' \
--attention_config.use_fp4_indexer_cache=True \
--tokenizer-mode deepseek_v4 \
Expand Down
6 changes: 6 additions & 0 deletions perf-changelog.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3445,6 +3445,12 @@
- "Add 1k1k/8k1k minimax recipe set under benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m2.5/"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1641

- config-keys:
- dsv4-fp4-b200-vllm
description:
- "Enable EPLB for DEP configs"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1655

- config-keys:
- minimaxm2.5-fp8-gb300-dynamo-vllm
description:
Expand Down
Loading