[AMD/ROCM] atom minimaxm2.5 fp4 on mi355x

.github/configs/amd-master.yaml

@@ -555,6 +555,30 @@ minimaxm2.5-fp8-mi355x-atom:
     - { tp: 4, conc-start: 4, conc-end: 128 }
     - { tp: 8, ep: 8, conc-start: 32, conc-end: 256 }
 
+minimaxm2.5-fp4-mi355x-atom:
+  image: rocm/atom:rocm7.2.2_ubuntu24.04_py3.12_pytorch_release_2.10.0_atom0.1.2.post
+  model: amd/MiniMax-M2.5-MXFP4
+  model-prefix: minimaxm2.5
+  runner: mi355x
+  precision: fp4
+  framework: atom
+  multinode: false
+  seq-len-configs:
+  - isl: 1024
+    osl: 1024
+    search-space:
+    - { tp: 1, conc-start: 4, conc-end: 1024 }
+    - { tp: 2, conc-start: 4, conc-end: 1024 }
+    - { tp: 4, conc-start: 4, conc-end: 128 }
+    - { tp: 8, conc-start: 4, conc-end: 16 }
+  - isl: 8192
+    osl: 1024
+    search-space:
+    - { tp: 1, conc-start: 4, conc-end: 1024 }
+    - { tp: 2, conc-start: 4, conc-end: 1024 }
+    - { tp: 4, conc-start: 4, conc-end: 128 }
+    - { tp: 8, conc-start: 4, conc-end: 16 }
+
 minimaxm2.5-fp4-mi355x-vllm:
   image: vllm/vllm-openai-rocm:v0.19.1
   model: amd/MiniMax-M2.5-MXFP4

benchmarks/single_node/minimaxm2.5_fp4_mi355x_atom.sh

@@ -0,0 +1,80 @@
+#!/usr/bin/env bash
+
+source "$(dirname "$0")/../benchmark_lib.sh"
+
+check_env_vars \
+    MODEL \
+    TP \
+    CONC \
+    ISL \
+    OSL \
+    RANDOM_RANGE_RATIO \
+    RESULT_FILENAME \
+    EP_SIZE \
+    DP_ATTENTION
+
+if [[ -n "$SLURM_JOB_ID" ]]; then
+  echo "JOB $SLURM_JOB_ID running on $SLURMD_NODENAME"
+fi
+
+echo "TP: $TP, CONC: $CONC, ISL: $ISL, OSL: $OSL, EP_SIZE: $EP_SIZE, DP_ATTENTION: $DP_ATTENTION"
+
+SERVER_LOG=/workspace/server.log
+PORT=${PORT:-8888}
+
+export OMP_NUM_THREADS=1
+
+# Calculate max-model-len based on ISL and OSL
+if [ "$ISL" = "1024" ] && [ "$OSL" = "1024" ]; then
+    CALCULATED_MAX_MODEL_LEN=""
+else
+    CALCULATED_MAX_MODEL_LEN=" --max-model-len 10240 "
+fi
+
+if [ "$EP_SIZE" -gt 1 ]; then
+  EP=" --enable-expert-parallel"
+else
+  EP=" "
+fi
+
+# Start GPU monitoring (power, temperature, clocks every second)
+start_gpu_monitor
+
+set -x
+
+python3 -m atom.entrypoints.openai_server \
+    --model $MODEL \
+    --server-port $PORT \
+    -tp $TP \
+    --kv_cache_dtype fp8 $CALCULATED_MAX_MODEL_LEN $EP \
+    --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 "$SERVER_PID"
+
+export PYTHONDONTWRITEBYTECODE=1
+run_benchmark_serving \
+    --model "$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 "$RESULT_FILENAME" \
+    --result-dir /workspace/ \
+    --trust-remote-code
+
+# After throughput, run evaluation only if RUN_EVAL is true
+if [ "${RUN_EVAL}" = "true" ]; then
+    run_eval --framework lm-eval --port "$PORT" 
+    append_lm_eval_summary
+fi
+
+# Stop GPU monitoring
+stop_gpu_monitor
+set +x

perf-changelog.yaml

@@ -1962,6 +1962,13 @@
     - "Drop --pipeline-parallel-size 1; keep --no-enable-prefix-caching and --max-cudagraph-capture-size 2048"
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1204
 
+- config-keys:
+    - minimaxm2.5-fp4-mi355x-atom
+  description:
+    - "Add MiniMax-M2.5 MXFP4 MI355X Atom benchmark (rocm/atom:rocm7.2.2_ubuntu24.04_py3.12_pytorch_release_2.10.0_atom0.1.2.post)"
+    - "Single-node sweep: TP1–TP8, 1k/1k and 8k/1k ISL/OSL"
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1042
+
 - config-keys:
     - dsv4-fp4-gb200-dynamo-vllm
   description: