diff --git a/perf-changelog.yaml b/perf-changelog.yaml index af01516c1..39184d7e2 100644 --- a/perf-changelog.yaml +++ b/perf-changelog.yaml @@ -4107,3 +4107,11 @@ - "Use the dedicated ARM64 MiniMax-M3 performance image; benchmark settings unchanged" - "Allocate FlashInfer MNNVL workspace for one-shot TP8 all-reduce during CUDA graph capture" pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1888 + +- config-keys: + - minimaxm3-fp8-gb300-dynamo-vllm + description: + - "Update the GB300 MiniMax-M3 Dynamo-vLLM image to vllm/vllm-openai:minimax-m3-perf-arm64-13.0.1-7a67223" + - "Use the dedicated ARM64 MiniMax-M3 performance image; benchmark settings unchanged" + - "Allocate FlashInfer MNNVL workspace for one-shot TP8 all-reduce during CUDA graph capture" + pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1897