From c69821587069d933e6a881076ccaf8cff19d5db4 Mon Sep 17 00:00:00 2001
From: functionstackx <47992694+functionstackx@users.noreply.github.com>
Date: Fri, 17 Apr 2026 22:48:04 -0400
Subject: [PATCH] Add MI355X config: glm5-fp8-sglang-mtp

Mirrors the existing glm5-fp8-mi355x-sglang non-MTP recipe and adds
EAGLE speculative decoding (num-steps=3, eagle-topk=1, num-draft-tokens=4)
via the standard spec-decoding=mtp suffix. SGLANG_ENABLE_SPEC_V2=1 is
set before launching the server as required for GLM-5 MTP. Script also
passes --use-chat-template to run_benchmark_serving, as required by
AGENTS.md for all MTP configs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
---
 .github/configs/amd-master.yaml               | 18 ++++
 benchmarks/single_node/glm5_fp8_mi355x_mtp.sh | 87 +++++++++++++++++++
 perf-changelog.yaml                           | 10 +++
 3 files changed, 115 insertions(+)
 create mode 100755 benchmarks/single_node/glm5_fp8_mi355x_mtp.sh

diff --git a/.github/configs/amd-master.yaml b/.github/configs/amd-master.yaml
index 2f8293688..993a075bd 100644
--- a/.github/configs/amd-master.yaml
+++ b/.github/configs/amd-master.yaml
@@ -301,6 +301,24 @@ glm5-fp8-mi355x-sglang:
     search-space:
     - { tp: 8, conc-start: 4, conc-end: 64 }
 
+glm5-fp8-mi355x-sglang-mtp:
+  image: lmsysorg/sglang-rocm:v0.5.10rc0-rocm720-mi35x-20260413
+  model: zai-org/GLM-5-FP8
+  model-prefix: glm5
+  runner: mi355x
+  precision: fp8
+  framework: sglang
+  multinode: false
+  seq-len-configs:
+  - isl: 1024
+    osl: 1024
+    search-space:
+    - { tp: 8, conc-start: 4, conc-end: 64, spec-decoding: mtp }
+  - isl: 8192
+    osl: 1024
+    search-space:
+    - { tp: 8, conc-start: 4, conc-end: 64, spec-decoding: mtp }
+
 glm5-fp8-mi355x-atom:
   image: rocm/atom:rocm7.2.1-ubuntu24.04-pytorch2.9.1-atom0.1.2.post
   model: zai-org/GLM-5-FP8
diff --git a/benchmarks/single_node/glm5_fp8_mi355x_mtp.sh b/benchmarks/single_node/glm5_fp8_mi355x_mtp.sh
new file mode 100755
index 000000000..f4b899011
--- /dev/null
+++ b/benchmarks/single_node/glm5_fp8_mi355x_mtp.sh
@@ -0,0 +1,87 @@
+#!/usr/bin/env bash
+
+source "$(dirname "$0")/../benchmark_lib.sh"
+
+check_env_vars \
+    MODEL \
+    TP \
+    CONC \
+    ISL \
+    OSL \
+    RANDOM_RANGE_RATIO \
+    RESULT_FILENAME
+
+if [[ -n "$SLURM_JOB_ID" ]]; then
+  echo "JOB $SLURM_JOB_ID running on $SLURMD_NODENAME"
+fi
+
+# GLM-5 requires transformers with glm_moe_dsa model type support.
+# However, the Image rocm/sgl-dev:v0.5.8.post1-rocm720-mi35x-20260219 doesn't provide this support.
+python3 -m pip install -U --no-cache-dir \
+  "git+https://github.com/huggingface/transformers.git@6ed9ee36f608fd145168377345bfc4a5de12e1e2"
+
+hf download "$MODEL"
+
+# ROCm / SGLang performance tuning for MI355X
+export SGLANG_ROCM_FUSED_DECODE_MLA=0
+export ROCM_QUICK_REDUCE_QUANTIZATION=INT4
+export SAFETENSORS_FAST_GPU=1
+export SGLANG_ENABLE_SPEC_V2=1
+
+SERVER_LOG=/workspace/server.log
+PORT=${PORT:-8888}
+
+EVAL_CONTEXT_ARGS=""
+if [ "${EVAL_ONLY}" = "true" ]; then
+    setup_eval_context
+    EVAL_CONTEXT_ARGS="--context-length $EVAL_MAX_MODEL_LEN"
+fi
+# Start GPU monitoring (power, temperature, clocks every second)
+start_gpu_monitor
+
+python3 -m sglang.launch_server \
+    --model-path $MODEL \
+    --host=0.0.0.0 \
+    --port $PORT \
+    --tensor-parallel-size $TP \
+    --trust-remote-code \
+    --tool-call-parser glm47 \
+    --reasoning-parser glm45 \
+    --mem-fraction-static 0.85 \
+    --model-loader-extra-config '{"enable_multithread_load": true, "num_threads": 8}' \
+    --nsa-prefill-backend tilelang \
+    --nsa-decode-backend tilelang $EVAL_CONTEXT_ARGS  \
+    --kv-cache-dtype fp8_e4m3 \
+    --speculative-algorithm EAGLE \
+    --speculative-num-steps 3 \
+    --speculative-eagle-topk 1 \
+    --speculative-num-draft-tokens 4 \
+    --disable-radix-cache> $SERVER_LOG 2>&1 &
+
+SERVER_PID=$!
+
+# Wait for server to be ready
+wait_for_server_ready --port "$PORT" --server-log "$SERVER_LOG" --server-pid "$SERVER_PID"
+
+run_benchmark_serving \
+    --model "$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 "$RESULT_FILENAME" \
+    --result-dir /workspace/ \
+    --use-chat-template
+
+# After throughput, run evaluation only if RUN_EVAL is true
+if [ "${RUN_EVAL}" = "true" ]; then
+    run_eval --framework lm-eval --port "$PORT"
+    append_lm_eval_summary
+fi
+
+# Stop GPU monitoring
+stop_gpu_monitor
+set +x
diff --git a/perf-changelog.yaml b/perf-changelog.yaml
index b4a289dbe..24f2e3a2c 100644
--- a/perf-changelog.yaml
+++ b/perf-changelog.yaml
@@ -1576,3 +1576,13 @@
     - "Mirrors the qwen3.5-fp8-mi355x-sglang non-MTP recipe and adds EAGLE speculative decoding (num-steps=3, eagle-topk=1, num-draft-tokens=4)"
     - "Configs: 1k1k (TP8/EP1, TP8/EP8, TP2/EP2) and 8k1k (TP2/EP2, TP4/EP1) with spec-decoding=mtp"
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXXX
+
+- config-keys:
+    - glm5-fp8-mi355x-sglang-mtp
+  description:
+    - "Add GLM-5 FP8 MI355X SGLang MTP benchmark"
+    - "Image: lmsysorg/sglang-rocm:v0.5.10rc0-rocm720-mi35x-20260413"
+    - "Model: zai-org/GLM-5-FP8"
+    - "Mirrors the glm5-fp8-mi355x-sglang non-MTP recipe and adds EAGLE speculative decoding (num-steps=3, eagle-topk=1, num-draft-tokens=4) behind SGLANG_ENABLE_SPEC_V2=1"
+    - "Configs: 1k1k and 8k1k, TP=8 conc 4-64 with spec-decoding=mtp"
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXXX