From 22012da3726e13217c9f5fc10610a53506f7a45f Mon Sep 17 00:00:00 2001
From: "claude[bot]" <41898282+claude[bot]@users.noreply.github.com>
Date: Sat, 2 May 2026 19:07:38 +0000
Subject: [PATCH] Update minimaxm2.5-fp8-mi355x-atom config

- Update Atom image to rocm7.2.2_ubuntu24.04_py3.12_pytorch_release_2.10.0_atom0.1.2.post
- Expand conc-end to 256 for tp2/tp4, remove tp8/ep8 configs
- Add gpu-memory-utilization 0.9 to benchmark script
- Add perf-changelog entry

Co-authored-by: functionstackx <functionstackx@users.noreply.github.com>
---
 .github/configs/amd-master.yaml                      | 12 +++++-------
 .../single_node/minimaxm2.5_fp8_mi355x_atom.sh       |  2 ++
 perf-changelog.yaml                                  |  7 +++++++
 3 files changed, 14 insertions(+), 7 deletions(-)

diff --git a/.github/configs/amd-master.yaml b/.github/configs/amd-master.yaml
index 949a8a106..7724538d8 100644
--- a/.github/configs/amd-master.yaml
+++ b/.github/configs/amd-master.yaml
@@ -591,7 +591,7 @@ minimaxm2.5-fp8-mi355x-vllm:
       - { tp: 8, ep: 8, conc-start: 2, conc-end: 2 }
 
 minimaxm2.5-fp8-mi355x-atom:
-  image: rocm/atom:rocm7.2.1-ubuntu24.04-pytorch2.9.1-atom0.1.2
+  image: rocm/atom:rocm7.2.2_ubuntu24.04_py3.12_pytorch_release_2.10.0_atom0.1.2.post
   model: MiniMaxAI/MiniMax-M2.5
   model-prefix: minimaxm2.5
   runner: mi355x
@@ -603,15 +603,13 @@ minimaxm2.5-fp8-mi355x-atom:
     - isl: 1024
       osl: 1024
       search-space:
-      - { tp: 2, conc-start: 4, conc-end: 128 }
-      - { tp: 4, conc-start: 4, conc-end: 128 }
-      - { tp: 8, ep: 8, conc-start: 32, conc-end: 256 }
+      - { tp: 2, conc-start: 4, conc-end: 256 }
+      - { tp: 4, conc-start: 4, conc-end: 256 }
     - isl: 8192
       osl: 1024
       search-space:
-      - { tp: 2, conc-start: 4, conc-end: 128 }
-      - { tp: 4, conc-start: 4, conc-end: 128 }
-      - { tp: 8, ep: 8, conc-start: 32, conc-end: 256 }
+      - { tp: 2, conc-start: 4, conc-end: 256 }
+      - { tp: 4, conc-start: 4, 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
diff --git a/benchmarks/single_node/minimaxm2.5_fp8_mi355x_atom.sh b/benchmarks/single_node/minimaxm2.5_fp8_mi355x_atom.sh
index ca84f8228..2a8c67da0 100755
--- a/benchmarks/single_node/minimaxm2.5_fp8_mi355x_atom.sh
+++ b/benchmarks/single_node/minimaxm2.5_fp8_mi355x_atom.sh
@@ -39,6 +39,7 @@ fi
 
 # Start GPU monitoring (power, temperature, clocks every second)
 start_gpu_monitor
+MEM_FRAC_STATIC=0.9
 
 set -x
 
@@ -47,6 +48,7 @@ python3 -m atom.entrypoints.openai_server \
     --server-port $PORT \
     -tp $TP \
     --kv_cache_dtype fp8 $CALCULATED_MAX_MODEL_LEN $EP \
+    --gpu-memory-utilization $MEM_FRAC_STATIC \
     --trust-remote-code \
     > $SERVER_LOG 2>&1 &
 
diff --git a/perf-changelog.yaml b/perf-changelog.yaml
index f50347c65..29b365007 100644
--- a/perf-changelog.yaml
+++ b/perf-changelog.yaml
@@ -2093,3 +2093,10 @@
     - "Dynamic scheduler-recv-interval: 30 for CONC>4, 10 otherwise"
     - "Remove --max-running-requests, reduce prefill/chunked from 81920 to 16384"
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1018
+
+- config-keys:
+    - minimaxm2.5-fp8-mi355x-atom
+  description:
+    - "Update Atom image to rocm/atom:rocm7.2.2_ubuntu24.04_py3.12_pytorch_release_2.10.0_atom0.1.2.post"
+    - "Search-space: expand conc-end to 256 for tp2/tp4 (1k/1k and 8k/1k); remove tp8/ep8 configs"
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1194