[Klaud Cold] Update GLM-5.2 FP8 B200 SGLang AgentX MTP image to nightly-dev-20261026-8252e3af / [Klaud Cold] 将 GLM-5.2 FP8 B200 SGLang AgentX MTP 镜像更新至 nightly-dev-20261026-8252e3af

inferencex-e2e/benchmarks/single_node/srt-slurm-recipes/glm5.2/sglang/b200-fp8-mtp/agentic.yaml

@@ -144,7 +144,7 @@ override_c16:
         dcp-size: 8
         chunked-prefill-size: 16384
         max-prefill-tokens: 16384
-        mem-fraction-static: 0.88
+        mem-fraction-static: 0.83
         max-running-requests: 32
         cuda-graph-max-bs-decode: 32
         # Hold prefills until min(0.5 x running, prefill-max-requests) are queued,

inferencex-e2e/configs/nvidia-master.yaml

@@ -5018,7 +5018,7 @@ glm5.2-fp4-b200-sglang-agentic-mtp:
       - { tp: 8, spec-decoding: mtp, kv-offloading: dram, kv-offload-backend: { name: hicache }, conc-list: [1, 4, 8, 12, 16], srt-recipe: agentic.yaml }
 
 glm5.2-fp8-b200-sglang-agentic-mtp:
-  image: lmsysorg/sglang:nightly-dev-20261020-ab580b2e@sha256:0af411bb320a95bb5e1c58b4c20c9852920a19aa9f12f148f69731b0dca3b576
+  image: lmsysorg/sglang:nightly-dev-20261026-8252e3af@sha256:b4e4abb9e1dc038f7e14c85805be97e4f74abb46e541bd4edd31401d50a2e593
   model: zai-org/GLM-5.2-FP8
   model-prefix: glm5.2
   runner: cluster:b200-nscale

inferencex-e2e/perf-changelog.yaml

@@ -11954,3 +11954,10 @@
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/4726
   scenario-type: [agentic-coding]
   append-only: true
+
+- config-keys:
+    - glm5.2-fp8-b200-sglang-agentic-mtp
+  description:
+    - "Update the SGLang image to nightly-dev-20261026-8252e3af and lower c16 mem-fraction-static to 0.83."
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/4543
+  scenario-type: [agentic-coding]