From 709571166e8e92fd39d365d267a7ba0b4ca487ba Mon Sep 17 00:00:00 2001 From: "claude[bot]" <41898282+claude[bot]@users.noreply.github.com> Date: Tue, 27 Oct 2026 12:23:37 +0000 Subject: [PATCH 1/2] chore(amd): update DSv4 FP4 MI355X vLLM AgentX MTP image to nightly-5d35ce39 Co-Authored-By: Claude Opus 5.5 --- inferencex-e2e/configs/amd-master.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/inferencex-e2e/configs/amd-master.yaml b/inferencex-e2e/configs/amd-master.yaml index d75eb9218e..32833c3940 100644 --- a/inferencex-e2e/configs/amd-master.yaml +++ b/inferencex-e2e/configs/amd-master.yaml @@ -988,7 +988,7 @@ qwen3.5-fp8-mi325x-sglang-mtp: # DeepSeek-V4-Pro-0813 FP4 AgentX on one MI355X node using vLLM DSpark K6. # Keep the historical config key for changelog compatibility. dsv4-fp4-mi355x-vllm-agentic-mtp: - image: vllm/vllm-openai-rocm:nightly-e9757321527ca1ecd514c07c1418dd2c53da3d19 + image: vllm/vllm-openai-rocm:nightly-5d35ce3934a7804b298559d2d99eec33792e1eb9@sha256:0c030ba9e2e5fd2f45f2a048a1c8bb239d9674b63297166b79e258b4ce65c6cc model: deepseek-ai/DeepSeek-V4-Pro-0813 model-prefix: dsv4 runner: cluster:mi355x-amds From 29655ec519775eda2e9bcdfcd35263cb88626ba9 Mon Sep 17 00:00:00 2001 From: "claude[bot]" <41898282+claude[bot]@users.noreply.github.com> Date: Tue, 27 Oct 2026 14:56:33 +0000 Subject: [PATCH 2/2] chore(amd): add perf-changelog entry for DSv4 FP4 MI355X vLLM AgentX MTP nightly-5d35ce39 Co-Authored-By: Claude Opus 5.5 --- inferencex-e2e/perf-changelog.yaml | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/inferencex-e2e/perf-changelog.yaml b/inferencex-e2e/perf-changelog.yaml index fd894cc02f..d04eedd2b0 100644 --- a/inferencex-e2e/perf-changelog.yaml +++ b/inferencex-e2e/perf-changelog.yaml @@ -11413,3 +11413,11 @@ - Restore the MI355X MiniMax-M3 FP4 vLLM AgentX TP2/TP4 grid from run 34609443618 attempt 2 using SRT recipes. Update nightly-2a02f6efe319c885e3ccbcecde402e0028f9ec1e to nightly-df6603f0583375f244d1ac6d36265462b7490392 (2026-10-22, digest sha256:81899892d73777a4ae81e16f39e749169687f29249505eb4c56da83bd097d41c). - Enable upstream MiniMax-M3 monokernel at TP4 with Model Runner V2, AITER MoE, FP8 KV/indexer and block size 128; keep TP2 on ordinary decode. Preserve EAGLE3-GQA with three draft tokens and use automatic golden acceptance for throughput with real verification for evals. pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/4418 + +- config-keys: + - dsv4-fp4-mi355x-vllm-agentic-mtp + scenario-type: + - agentic-coding + description: + - Update the vLLM ROCm image from nightly-e9757321 to nightly-5d35ce39, pinned by digest. + pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/4578