Skip to content

Update MiniMax M2.5 FP8 H200 vLLM agg recipes #1354

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
Ankur-singh merged 2 commits into from
May 26, 2026
+35 −10
Merged

Update MiniMax M2.5 FP8 H200 vLLM agg recipes #1354

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
8e3037d
Tune MiniMax M2.5 FP8 H200 vLLM agg
anish-shanbhag May 7, 2026
111df5b
Merge branch 'main' into ashanbhag/minimax-h200
Ankur-singh May 26, 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
4 changes: 2 additions & 2 deletions .github/configs/nvidia-master.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4754,11 +4754,11 @@ minimaxm2.5-fp8-h200-vllm:
- isl: 1024
osl: 1024
search-space:
- { tp: 8, conc-start: 4, conc-end: 128 }
- { tp: 4, conc-start: 1, conc-end: 256 }
- isl: 8192
osl: 1024
search-space:
- { tp: 8, conc-start: 4, conc-end: 128 }
- { tp: 4, conc-start: 1, conc-end: 256 }

# Diverged from minimaxm2.5-fp8-h200-vllm (agentic-coding sibling). Metadata is
# identical to origin/main's minimaxm2.5-fp8-h200-vllm; the split exists because this
Expand Down
33 changes: 25 additions & 8 deletions benchmarks/single_node/minimaxm2.5_fp8_h200.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,23 +27,40 @@ if [ "${EVAL_ONLY}" = "true" ]; then
MAX_MODEL_LEN="$EVAL_MAX_MODEL_LEN"
fi

if [ "$EP_SIZE" -ge 1 ]; then
EP=" --enable-expert-parallel"
export PYTHONNOUSERSITE=1
export SAFETENSORS_FAST_GPU=1
export VLLM_USE_DEEP_GEMM=0
export VLLM_BLOCKSCALE_FP8_GEMM_FLASHINFER=0
export VLLM_FLOAT32_MATMUL_PRECISION=high

COMPILATION_CONFIG=${COMPILATION_CONFIG:-'{"mode":3,"cudagraph_mode":"PIECEWISE","pass_config":{"fuse_minimax_qk_norm":true}}'}
MAX_NUM_SEQS=${MAX_NUM_SEQS:-512}
MAX_NUM_BATCHED_TOKENS=${MAX_NUM_BATCHED_TOKENS:-32768}

if [ "$EP_SIZE" -gt 1 ]; then
EP=(--enable-expert-parallel)
else
EP=" "
EP=()
fi

# Start GPU monitoring (power, temperature, clocks every second)
start_gpu_monitor

set -x
vllm serve $MODEL --port $PORT \
--tensor-parallel-size=$TP \
$EP \
vllm serve "$MODEL" --port "$PORT" \
--tensor-parallel-size="$TP" \
"${EP[@]}" \
--gpu-memory-utilization 0.95 \
--max-model-len $MAX_MODEL_LEN \
--max-model-len "$MAX_MODEL_LEN" \
--max-num-seqs "$MAX_NUM_SEQS" \
--max-num-batched-tokens "$MAX_NUM_BATCHED_TOKENS" \
--kv-cache-dtype fp8 \
--moe-backend triton \
--attention-backend FLASHINFER \
--enable-flashinfer-autotune \
--compilation-config "$COMPILATION_CONFIG" \
--no-enable-prefix-caching \
--trust-remote-code > $SERVER_LOG 2>&1 &
--trust-remote-code > "$SERVER_LOG" 2>&1 &

SERVER_PID=$!

Expand Down
8 changes: 8 additions & 0 deletions perf-changelog.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3129,3 +3129,11 @@
description:
- "Add --use-chat-template to run_benchmark_serving so prompts are formatted with the Qwen chat template (matching the other Qwen MTP recipes)"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1555

- config-keys:
- minimaxm2.5-fp8-h200-vllm
description:
- "Update MiniMax-M2.5 FP8 H200 vLLM to vllm/vllm-openai:v0.20.1-ubuntu2404"
- "Set vLLM serving knobs in benchmarks/single_node/minimaxm2.5_fp8_h200.sh: generated benchmark max-model-len, previous eval max-model-len handling, fp8 KV cache, FlashInfer attention/autotune, Triton MoE, and MiniMax QK norm fusion"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1354