From 4a4f5749622734d2f8fd3e62af9a25557b85a238 Mon Sep 17 00:00:00 2001
From: "claude[bot]" <41898282+claude[bot]@users.noreply.github.com>
Date: Fri, 27 Mar 2026 07:06:09 +0000
Subject: [PATCH 1/2] mi325x: port kimik2.5 int4 changes from mi355x PR #950

- Upgrade vLLM ROCm image from v0.16.0 to v0.18.0
- Enable AITER MLA (export VLLM_ROCM_USE_AITER=1)
- Add --max-num-seqs 256
- Remove --disable-log-requests

Co-authored-by: functionstackx <functionstackx@users.noreply.github.com>
---
 .github/configs/amd-master.yaml                | 2 +-
 benchmarks/single_node/kimik2.5_int4_mi325x.sh | 3 ++-
 perf-changelog.yaml                            | 8 ++++++++
 3 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/.github/configs/amd-master.yaml b/.github/configs/amd-master.yaml
index fa48ee1be..b9560803e 100644
--- a/.github/configs/amd-master.yaml
+++ b/.github/configs/amd-master.yaml
@@ -316,7 +316,7 @@ kimik2.5-int4-mi355x-vllm:
     - { tp: 8, conc-start: 4, conc-end: 64 }
 
 kimik2.5-int4-mi325x-vllm:
-  image: vllm/vllm-openai-rocm:v0.16.0
+  image: vllm/vllm-openai-rocm:v0.18.0
   model: moonshotai/Kimi-K2.5
   model-prefix: kimik2.5
   runner: mi325x
diff --git a/benchmarks/single_node/kimik2.5_int4_mi325x.sh b/benchmarks/single_node/kimik2.5_int4_mi325x.sh
index e6b7629ea..8ca56dc29 100755
--- a/benchmarks/single_node/kimik2.5_int4_mi325x.sh
+++ b/benchmarks/single_node/kimik2.5_int4_mi325x.sh
@@ -33,13 +33,14 @@ PORT=${PORT:-8888}
 start_gpu_monitor
 
 set -x
+export VLLM_ROCM_USE_AITER=1
 vllm serve $MODEL --port $PORT \
 --tensor-parallel-size=$TP \
 --gpu-memory-utilization 0.95 \
 --max-model-len $MAX_MODEL_LEN \
 --block-size=64 \
---disable-log-requests \
 --trust-remote-code \
+--max-num-seqs 256 \
 --mm-encoder-tp-mode data > $SERVER_LOG 2>&1 &
 
 SERVER_PID=$!
diff --git a/perf-changelog.yaml b/perf-changelog.yaml
index b85245458..2f2f558f1 100644
--- a/perf-changelog.yaml
+++ b/perf-changelog.yaml
@@ -1085,3 +1085,11 @@
     - "Triton Fused Moe Tuning https://github.com/vllm-project/vllm/pull/35093"
     - "Add --max-num-seqs 256, remove --disable-log-requests"
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/950
+
+- config-keys:
+    - kimik2.5-int4-mi325x-vllm
+  description:
+    - "Upgrade vLLM ROCm image from v0.16.0 to v0.18.0"
+    - "Enable AITER MLA, export VLLM_ROCM_USE_AITER=1"
+    - "Add --max-num-seqs 256, remove --disable-log-requests"
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX

From 974e55516011357f7207e478825d00540d213208 Mon Sep 17 00:00:00 2001
From: functionstackx <47992694+functionstackx@users.noreply.github.com>
Date: Fri, 27 Mar 2026 03:10:08 -0400
Subject: [PATCH 2/2] Update perf-changelog.yaml

---
 perf-changelog.yaml | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/perf-changelog.yaml b/perf-changelog.yaml
index 2f2f558f1..633a25be2 100644
--- a/perf-changelog.yaml
+++ b/perf-changelog.yaml
@@ -1090,6 +1090,7 @@
     - kimik2.5-int4-mi325x-vllm
   description:
     - "Upgrade vLLM ROCm image from v0.16.0 to v0.18.0"
-    - "Enable AITER MLA, export VLLM_ROCM_USE_AITER=1"
+    - "Enable AITER MLA, export VLLM_ROCM_USE_AITER=1, https://github.com/vllm-project/vllm/issues/35641"
+    - "Triton Fused Moe Tuning https://github.com/vllm-project/vllm/pull/35093"
     - "Add --max-num-seqs 256, remove --disable-log-requests"
-  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/957