[WIP] [NV] Qwen3.5 FP8 H200 SGLang

.github/configs/nvidia-master.yaml

@@ -1974,6 +1974,28 @@ dsr1-fp8-h200-sglang:
     search-space:
     - { tp: 8, conc-start: 4, conc-end: 64 }
 
+qwen3.5-fp8-h200-sglang:
+  image: lmsysorg/sglang:v0.5.9-cu129-amd64
+  model: Qwen/Qwen3.5-397B-A17B-FP8
+  model-prefix: qwen3.5
+  runner: h200
+  precision: fp8
+  framework: sglang
+  multinode: false
+  seq-len-configs:
+  - isl: 1024
+    osl: 1024
+    search-space:
+    - { tp: 8, ep: 8, conc-start: 4, conc-end: 64 }
+  - isl: 1024
+    osl: 8192
+    search-space:
+    - { tp: 8, ep: 8, conc-start: 4, conc-end: 64 }
+  - isl: 8192
+    osl: 1024
+    search-space:
+    - { tp: 8, ep: 8, conc-start: 4, conc-end: 64 }
+
 dsr1-fp8-h200-trt:
   image: nvcr.io#nvidia/tensorrt-llm/release:1.1.0rc2.post2
   model: deepseek-ai/DeepSeek-R1-0528

benchmarks/single_node/qwen3.5_fp8_h200.sh

@@ -0,0 +1,79 @@
+#!/usr/bin/env bash
+
+source "$(dirname "$0")/../benchmark_lib.sh"
+
+check_env_vars \
+    MODEL \
+    TP \
+    CONC \
+    ISL \
+    OSL \
+    RANDOM_RANGE_RATIO \
+    RESULT_FILENAME \
+    EP_SIZE 
+
+if [[ -n "$SLURM_JOB_ID" ]]; then
+  echo "JOB $SLURM_JOB_ID running on $SLURMD_NODENAME"
+fi
+
+nvidia-smi
+
+hf download "$MODEL"
+
+SERVER_LOG=/workspace/server.log
+PORT=${PORT:-8888}
+MAX_SEQ_LEN=$((ISL + OSL + 20))
+
+echo "CONC: $CONC, ISL: $ISL, OSL: $OSL, MAX_SEQ_LEN: $MAX_SEQ_LEN"
+
+set -x
+python3 -m sglang.launch_server \
+  --model "$MODEL" \
+  --host 0.0.0.0 \
+  --port "$PORT" \
+  --tp "$TP" \
+  --expert-parallel-size "$EP_SIZE" \
+  --reasoning-parser qwen3 \
+  --tool-call-parser qwen3_coder \
+  --enable-flashinfer-allreduce-fusion \
+  --max-running-requests 128 \
+  --chunked-prefill-size 16384 \
+  --decode-log-interval 1 \
+  --mem-fraction-static 0.8 \
+  --cuda-graph-max-bs "$CONC" \
+  --context-length "$MAX_SEQ_LEN" \
+  --kv-cache-dtype fp8_e4m3 \
+  --quantization fp8 \
+  --attention-backend flashinfer \
+  --stream-interval 50 \
+  --tokenizer-worker-num 6 \
+  --mamba-ssm-dtype bfloat16 \
+  --disable-radix-cache \
+  --trust-remote-code \
+  > "$SERVER_LOG" 2>&1 &
+
+SERVER_PID=$!
+
+# Wait for server to be ready
+wait_for_server_ready --port "$PORT" --server-log "$SERVER_LOG" --server-pid "$SERVER_PID"
+
+pip install -q datasets pandas
+
+run_benchmark_serving \
+    --model "$MODEL" \
+    --port "$PORT" \
+    --backend vllm \
+    --input-len "$ISL" \
+    --output-len "$OSL" \
+    --random-range-ratio "$RANDOM_RANGE_RATIO" \
+    --num-prompts "$((CONC * 10))" \
+    --max-concurrency "$CONC" \
+    --result-filename "$RESULT_FILENAME" \
+    --result-dir /workspace/
+
+# After throughput, run evaluation only if RUN_EVAL is true
+if [ "${RUN_EVAL}" = "true" ]; then
+    run_eval --framework lm-eval --port "$PORT" --concurrent-requests $CONC
+    append_lm_eval_summary
+fi
+set +x

perf-changelog.yaml

@@ -834,7 +834,7 @@
     - "TP=8, concurrency 4-64 for 1k1k, 1k8k, and 8k1k sequence lengths"
     - "following https://docs.vllm.ai/projects/recipes/en/latest/moonshotai/Kimi-K2.5.html"
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/839
-
+  
 - config-keys:
     - dsr1-fp8-mi355x-sglang-disagg
     - dsr1-fp8-mi355x-sglang-disagg-mtp
@@ -865,3 +865,13 @@
     - "Uses trtllm_mha attention backend and flashinfer_trtllm MOE runner"
     - "Enable SGLANG_ENABLE_FLASHINFER_GEMM=true, NCCL_NVLS_ENABLE=1"
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/804
+
+- config-keys: 
+    - qwen3.5-fp8-h200-sglang
+  description:
+    - "Add Qwen 3.5 FP8 H200 SGLang configuration"
+    - "Model: Qwen/Qwen3.5-397B-A17B-FP8, runner: h200, image: lmsysorg/sglang:v0.5.8-cu130-amd64"
+    - "Benchmark script: benchmarks/single_node/qwen3.5_fp8_h200.sh"
+    - "Server: reasoning-parser qwen3, tool-call-parser qwen3_coder, enable-flashinfer-allreduce-fusion, mem-fraction-static 0.8"
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/855
+