Update dsr1-fp4-mi355x-sglang SGLang image to v0.5.12-rocm700-mi35x

.github/configs/amd-master.yaml

@@ -1,5 +1,5 @@
 dsr1-fp4-mi355x-sglang:
-  image: lmsysorg/sglang:v0.5.9-rocm700-mi35x
+  image: lmsysorg/sglang:v0.5.12-rocm700-mi35x
   model: amd/DeepSeek-R1-0528-MXFP4-Preview
   model-prefix: dsr1
   runner: mi355x
@@ -17,10 +17,14 @@ dsr1-fp4-mi355x-sglang:
       osl: 1024
       search-space:
       - { tp: 8, conc-start: 4, conc-end: 64 }
-    agentic-coding:
-    - duration: 1800
-      search-space:
-      - { tp: 8, offloading: none, conc-list: [1, 2, 4, 8, 12, 16, 32, 64, 128, 256] }
+    # Agentic-coding sweep commented out for this image-bump PR — the
+    # 10-conc agentic matrix amplifies sweep cost and the bump validation
+    # only needs the fixed-seq-len throughput shape. Re-enable once the
+    # bump merges; the next agentic cron PR will pick it up.
+    # agentic-coding:
+    # - duration: 1800
+    #   search-space:
+    #   - { tp: 8, offloading: none, conc-list: [1, 2, 4, 8, 12, 16, 32, 64, 128, 256] }
 
 dsr1-fp4-mi355x-atom:
   image: rocm/atom:rocm7.1.1-ubuntu24.04-pytorch2.9-atom0.1.1-MI350x

perf-changelog.yaml

@@ -2901,3 +2901,9 @@
   description:
     - "Add MI355X config: qwen3.5-fp4-sglang-mtp using lmsysorg/sglang-rocm:v0.5.12-rocm720-mi35x-20260517"
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1445
+
+- config-keys:
+    - dsr1-fp4-mi355x-sglang
+  description:
+    - "Update SGLang image to v0.5.12-rocm700-mi35x"
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1431