From a4d66a6e2dabf924f899adb4518e345d86bb19d5 Mon Sep 17 00:00:00 2001 From: Duyi-Wang Date: Tue, 27 Oct 2026 08:07:54 +0000 Subject: [PATCH 1/2] perf(amd): update MI355X DeepSeek V4 MoRI to ROCm 10 nightly MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 将 MI355X DeepSeek V4 Pro MoRI 分离式服务镜像更新为最新 ROCm 10 nightly，并追加性能变更记录。 --- inferencex-e2e/configs/amd-master.yaml | 2 +- inferencex-e2e/perf-changelog.yaml | 9 +++++++++ 2 files changed, 10 insertions(+), 1 deletion(-) diff --git a/inferencex-e2e/configs/amd-master.yaml b/inferencex-e2e/configs/amd-master.yaml index d75eb9218e..06068583c7 100644 --- a/inferencex-e2e/configs/amd-master.yaml +++ b/inferencex-e2e/configs/amd-master.yaml @@ -1197,7 +1197,7 @@ dsv41flash-fp4-mi325x-vllm-agentic-dspark: - { tp: 2, kv-offloading: none, spec-decoding: mtp, conc-list: [1, 2, 4, 8, 16, 32], srt-recipe: agentic.yaml } dsv4-fp4-mi355x-sglang-disagg-agentic-umbp-dspark: - image: lmsysorg/sglang-rocm:v0.5.22-rocm720-mi35x-20261019 + image: lmsysorg/sglang-rocm:v0.5.22-rocm10-mi35x-20261026 model: deepseek-ai/DeepSeek-V4-Pro-0813 model-prefix: dsv4 runner: cluster:mi355x-amds diff --git a/inferencex-e2e/perf-changelog.yaml b/inferencex-e2e/perf-changelog.yaml index fd894cc02f..2138030406 100644 --- a/inferencex-e2e/perf-changelog.yaml +++ b/inferencex-e2e/perf-changelog.yaml @@ -11413,3 +11413,12 @@ - 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-sglang-disagg-agentic-umbp-dspark + scenario-type: + - agentic-coding + description: + - "Update the MI355X DeepSeek-V4-Pro-0813 SGLang MoRI disaggregated AgentX serving image from lmsysorg/sglang-rocm:v0.5.22-rocm720-mi35x-20261019 to lmsysorg/sglang-rocm:v0.5.22-rocm10-mi35x-20261026, the latest published MI35x ROCm 10 nightly (digest sha256:edabad83728ab04e3eae39f315eb16261eeb64dbdb051112bd22fc88560b60c4)." + - "Preserve all eight concurrency points (1, 16, 32, 48, 192, 256, 384, 512), prefill/decode topology, DSpark settings, UMBP offload, MoRI transfer, and benchmark client image. The workload binder supplies the serving image to every selected recipe variant." + pr-link: TBD From 403fcf148b930538ea675a8008fc00d733e1a214 Mon Sep 17 00:00:00 2001 From: Duyi-Wang Date: Tue, 27 Oct 2026 08:08:55 +0000 Subject: [PATCH 2/2] chore: link MI355X image refresh changelog to PR MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 为 MI355X 镜像更新的性能变更记录补充 PR 链接。 --- inferencex-e2e/perf-changelog.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/inferencex-e2e/perf-changelog.yaml b/inferencex-e2e/perf-changelog.yaml index 2138030406..9e0f31add7 100644 --- a/inferencex-e2e/perf-changelog.yaml +++ b/inferencex-e2e/perf-changelog.yaml @@ -11421,4 +11421,4 @@ description: - "Update the MI355X DeepSeek-V4-Pro-0813 SGLang MoRI disaggregated AgentX serving image from lmsysorg/sglang-rocm:v0.5.22-rocm720-mi35x-20261019 to lmsysorg/sglang-rocm:v0.5.22-rocm10-mi35x-20261026, the latest published MI35x ROCm 10 nightly (digest sha256:edabad83728ab04e3eae39f315eb16261eeb64dbdb051112bd22fc88560b60c4)." - "Preserve all eight concurrency points (1, 16, 32, 48, 192, 256, 384, 512), prefill/decode topology, DSpark settings, UMBP offload, MoRI transfer, and benchmark client image. The workload binder supplies the serving image to every selected recipe variant." - pr-link: TBD + pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/4571