From bdfd4f863175ae77ef907e08d1c1e6fec4187ea7 Mon Sep 17 00:00:00 2001 From: functionstackx <47992694+functionstackx@users.noreply.github.com> Date: Fri, 4 Sep 2026 18:05:31 -0400 Subject: [PATCH 1/2] feat(config): extend GLM-5.2 B300 SGLang AgentX curve to concurrency 48 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add concurrency points 20, 24, 28, 32, 40, and 48 to the TP8 HiCache MTP arm of glm5.2-fp4-b300-sglang-agentic-mtp, and move the recipe from the retired cluster:b300-nv fleet to cluster:b300-dsxe so the sweep can be generated and scheduled again. 为 glm5.2-fp4-b300-sglang-agentic-mtp 的 TP8 HiCache MTP 分支新增并发度 20、24、28、32、40、48，并将配方从已下线的 cluster:b300-nv 迁移到 cluster:b300-dsxe，使 sweep 能重新生成与调度。 Co-Authored-By: Claude Fable 5.1 --- configs/nvidia-master.yaml | 4 ++-- perf-changelog.yaml | 9 +++++++++ 2 files changed, 11 insertions(+), 2 deletions(-) diff --git a/configs/nvidia-master.yaml b/configs/nvidia-master.yaml index 0a6c86df6..8c880b8a8 100644 --- a/configs/nvidia-master.yaml +++ b/configs/nvidia-master.yaml @@ -8991,7 +8991,7 @@ glm5.2-fp4-b300-sglang-agentic-mtp: image: lmsysorg/sglang:v0.5.16-cu130 model: nvidia/GLM-5.2-NVFP4 model-prefix: glm5.2 - runner: cluster:b300-nv + runner: cluster:b300-dsxe precision: fp4 framework: sglang multinode: false @@ -8999,7 +8999,7 @@ glm5.2-fp4-b300-sglang-agentic-mtp: agentic-coding: - dram-utilization: 0.80 search-space: - - { tp: 8, spec-decoding: mtp, kv-offloading: dram, kv-offload-backend: { name: hicache }, conc-list: [1, 4, 8, 12, 16] } + - { tp: 8, spec-decoding: mtp, kv-offloading: dram, kv-offload-backend: { name: hicache }, conc-list: [1, 4, 8, 12, 16, 20, 24, 28, 32, 40, 48] } # First GLM-5.2 AgentX recipe on B200, shipped spec-decode-only per the AgentX # policy that agentic arms enable speculative decoding rather than running a diff --git a/perf-changelog.yaml b/perf-changelog.yaml index 14434c76a..482045d9d 100644 --- a/perf-changelog.yaml +++ b/perf-changelog.yaml @@ -6878,3 +6878,12 @@ description: - "Refresh the Kimi-K3 GB300 AgentX configurations with updated configs." pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2811 + +- config-keys: + - glm5.2-fp4-b300-sglang-agentic-mtp + scenario-type: + - agentic-coding + description: + - "Extend the GLM-5.2 NVFP4 B300 SGLang AgentX MTP TP8 HiCache curve from concurrency 16 to 48 with new points at 20, 24, 28, 32, 40, and 48." + - "Move the recipe from the retired cluster:b300-nv fleet to cluster:b300-dsxe so the sweep can be generated and scheduled again." + pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX From 778a004c364065eafafbc13f9e2e4d340fe59277 Mon Sep 17 00:00:00 2001 From: functionstackx <47992694+functionstackx@users.noreply.github.com> Date: Fri, 4 Sep 2026 18:08:29 -0400 Subject: [PATCH 2/2] chore(changelog): fill the PR link for #2829 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 补全 perf-changelog 条目中的 PR 链接（#2829）。 Co-Authored-By: Claude Fable 5.1 --- perf-changelog.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/perf-changelog.yaml b/perf-changelog.yaml index 482045d9d..5386da981 100644 --- a/perf-changelog.yaml +++ b/perf-changelog.yaml @@ -6886,4 +6886,4 @@ description: - "Extend the GLM-5.2 NVFP4 B300 SGLang AgentX MTP TP8 HiCache curve from concurrency 16 to 48 with new points at 20, 24, 28, 32, 40, and 48." - "Move the recipe from the retired cluster:b300-nv fleet to cluster:b300-dsxe so the sweep can be generated and scheduled again." - pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX + pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2829