From 937d423e844b55f9c2ad6889ce99c5f758f3d166 Mon Sep 17 00:00:00 2001 From: Klaud-Cold Date: Thu, 10 Sep 2026 00:46:06 +0000 Subject: [PATCH 1/2] perf(minimaxm3): pin H200 vLLM AgentX MTP image to v0.29.0 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Update minimaxm3-fp8-h200-vllm-agentic-mtp from vllm/vllm-openai:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36 to vllm/vllm-openai:v0.29.0 pinned by manifest digest sha256:c2914767605584b6d8f45686b82de173ecc99e781897aa3d0a66dacd72c51ae1. The recipe script, model, precision, TP8 topology, EAGLE3 settings, workloads and the c1-c14 grid are unchanged. 将 minimaxm3-fp8-h200-vllm-agentic-mtp 的 vLLM 镜像从 vllm/vllm-openai:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36 更新为按清单摘要固定的 vllm/vllm-openai:v0.29.0 （sha256:c2914767605584b6d8f45686b82de173ecc99e781897aa3d0a66dacd72c51ae1）。 配方脚本、模型、精度、TP8 拓扑、EAGLE3 设置、工作负载与 c1-c14 网格均未改动。 Co-Authored-By: Claude Fable 5.1 --- configs/nvidia-master.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/configs/nvidia-master.yaml b/configs/nvidia-master.yaml index 0c44ca1af5..3df4c7ec95 100644 --- a/configs/nvidia-master.yaml +++ b/configs/nvidia-master.yaml @@ -7649,7 +7649,7 @@ minimaxm3-fp8-h100-vllm-agentic-mtp: - { tp: 8, spec-decoding: mtp, kv-offloading: dram, kv-offload-backend: { name: mooncake, version: "0.3.11.post1" }, conc-list: [6, 8] } minimaxm3-fp8-h200-vllm-agentic-mtp: - image: vllm/vllm-openai:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36 + image: vllm/vllm-openai:v0.29.0@sha256:c2914767605584b6d8f45686b82de173ecc99e781897aa3d0a66dacd72c51ae1 model: MiniMaxAI/MiniMax-M3-MXFP8 model-prefix: minimaxm3 runner: cluster:h200-dgxc From 7568b8f8529b843ba13dcfa45dfe1c43d49ae23f Mon Sep 17 00:00:00 2001 From: Klaud-Cold Date: Thu, 10 Sep 2026 02:31:20 +0000 Subject: [PATCH 2/2] docs(changelog): record minimaxm3 H200 vLLM v0.29.0 image refresh MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Append the perf-changelog entry for minimaxm3-fp8-h200-vllm-agentic-mtp (vllm/vllm-openai:v0.29.0, digest-pinned) linking PR #2953. 为 minimaxm3-fp8-h200-vllm-agentic-mtp 追加 perf-changelog 条目 （vllm/vllm-openai:v0.29.0，按摘要固定），关联 PR #2953。 Co-Authored-By: Claude Fable 5.1 --- perf-changelog.yaml | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/perf-changelog.yaml b/perf-changelog.yaml index e4f1ab141c..1ec62017eb 100644 --- a/perf-changelog.yaml +++ b/perf-changelog.yaml @@ -7021,3 +7021,9 @@ - "Refresh the full GB200 Dynamo-vLLM AgentX configuration across the configured aggregate and disaggregated topology points." - "Retain aggregate SimpleCPU-offload concurrency 20 and 30, and configure Mooncake host KV storage, NIXL over UCX, explicit thinking mode, and the committed MiniMax-M3 EAGLE3-GQA acceptance target." pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2807 + +- config-keys: + - minimaxm3-fp8-h200-vllm-agentic-mtp + description: + - "Update vLLM image from vllm/vllm-openai:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36 (2026-09-07 upstream nightly, tag commit vllm-project/vllm@d9105ea8) to vllm/vllm-openai:v0.29.0 pinned by manifest digest sha256:c2914767605584b6d8f45686b82de173ecc99e781897aa3d0a66dacd72c51ae1 (vLLM release v0.29.0, tag commit vllm-project/vllm@98dff2a8, Docker Hub tag pushed 2026-09-09T06:06Z, CUDA 13.0 build). benchmarks/single_node/agentic/minimaxm3_fp8_h200_mtp.sh is unchanged: TRITON_ATTN attention, fp8 KV, EAGLE3 with the Inferact MiniMax-M3 EAGLE3-GQA draft and the committed golden synthetic acceptance length 2.78, Mooncake 0.3.11.post1 DRAM offload on the host-tier arm; the resident TP8 c1/c2/c4/c6/c8/c10 and Mooncake DRAM offload c12/c14 grid is unchanged." + pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2953