From 6cb47f7db429287cabd0a2a338091fab82a2d481 Mon Sep 17 00:00:00 2001 From: Ankur-singh Date: Tue, 7 Apr 2026 10:22:43 -0700 Subject: [PATCH 1/5] Update minimax fp8 b200 configs --- .github/configs/nvidia-master.yaml | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/.github/configs/nvidia-master.yaml b/.github/configs/nvidia-master.yaml index b039f775c..00e6994f2 100644 --- a/.github/configs/nvidia-master.yaml +++ b/.github/configs/nvidia-master.yaml @@ -3130,15 +3130,15 @@ minimaxm2.5-fp8-b200-vllm: - isl: 1024 osl: 1024 search-space: - - { tp: 2, conc-start: 4, conc-end: 512 } - - { tp: 2, ep: 2, conc-start: 4, conc-end: 256 } - - { tp: 4, conc-start: 4, conc-end: 512 } - - { tp: 4, ep: 4, conc-start: 16, conc-end: 64 } + - { tp: 2, ep: 2, conc-start: 512, conc-end: 512 } + - { tp: 4, conc-start: 4, conc-end: 128 } + - { tp: 4, ep: 4, conc-start: 256, conc-end: 512 } - isl: 8192 osl: 1024 search-space: - - { tp: 2, conc-start: 4, conc-end: 256 } - - { tp: 4, conc-start: 4, conc-end: 256 } + - { tp: 2, conc-start: 64, conc-end: 512 } + - { tp: 4, conc-start: 4, conc-end: 64 } + - { tp: 4, conc-start: 512, conc-end: 512 } gptoss-fp4-h100-vllm: image: vllm/vllm-openai:v0.18.0 From 01a451ea064369bcbaea467a4eab5a022e570e90 Mon Sep 17 00:00:00 2001 From: Ankur-singh Date: Tue, 7 Apr 2026 10:25:00 -0700 Subject: [PATCH 2/5] Add perf-changelog entry for minimaxm2.5-fp8-b200-vllm update Update perf-changelog.yaml --- perf-changelog.yaml | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/perf-changelog.yaml b/perf-changelog.yaml index f7f824b32..0f36e41f5 100644 --- a/perf-changelog.yaml +++ b/perf-changelog.yaml @@ -1265,3 +1265,10 @@ description: - "Add Qwen3.5-397B-A17B-FP8 H200 SGLang MTP (EAGLE speculative decoding)" pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1001 + +- config-keys: + - minimaxm2.5-fp8-b200-vllm + description: + - "Update MiniMax-M2.5 FP8 B200 config with new search spaces and vllm args" + pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/xxxx + From 93adcd0d841be53bbee7d7a1709ceff442822254 Mon Sep 17 00:00:00 2001 From: "claude[bot]" <41898282+claude[bot]@users.noreply.github.com> Date: Tue, 7 Apr 2026 17:32:43 +0000 Subject: [PATCH 3/5] Fix perf-changelog PR link to #1010 Co-authored-by: Ankur Singh --- perf-changelog.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/perf-changelog.yaml b/perf-changelog.yaml index 0f36e41f5..507d2d449 100644 --- a/perf-changelog.yaml +++ b/perf-changelog.yaml @@ -1270,5 +1270,5 @@ - minimaxm2.5-fp8-b200-vllm description: - "Update MiniMax-M2.5 FP8 B200 config with new search spaces and vllm args" - pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/xxxx + pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1010 From bcc899fa49e8179f15042fe9a68bd39dd0e3956e Mon Sep 17 00:00:00 2001 From: Ankur-singh Date: Tue, 7 Apr 2026 12:22:52 -0700 Subject: [PATCH 4/5] Update minimax config --- .github/configs/nvidia-master.yaml | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/.github/configs/nvidia-master.yaml b/.github/configs/nvidia-master.yaml index 00e6994f2..9a6e1b01c 100644 --- a/.github/configs/nvidia-master.yaml +++ b/.github/configs/nvidia-master.yaml @@ -3130,15 +3130,15 @@ minimaxm2.5-fp8-b200-vllm: - isl: 1024 osl: 1024 search-space: + - { tp: 2, conc-start: 4, conc-end: 512 } + - { tp: 4, conc-start: 4, conc-end: 512 } - { tp: 2, ep: 2, conc-start: 512, conc-end: 512 } - - { tp: 4, conc-start: 4, conc-end: 128 } - { tp: 4, ep: 4, conc-start: 256, conc-end: 512 } - isl: 8192 osl: 1024 search-space: - - { tp: 2, conc-start: 64, conc-end: 512 } - - { tp: 4, conc-start: 4, conc-end: 64 } - - { tp: 4, conc-start: 512, conc-end: 512 } + - { tp: 2, conc-start: 4, conc-end: 512 } + - { tp: 4, conc-start: 4, conc-end: 512 } gptoss-fp4-h100-vllm: image: vllm/vllm-openai:v0.18.0 From 6d7481c144109b1f842d2f364efe0acbf2ca17a7 Mon Sep 17 00:00:00 2001 From: hshrivastava-droid Date: Tue, 7 Apr 2026 16:59:24 -0700 Subject: [PATCH 5/5] Update description for MiniMax-M2.5 FP8 B200 config --- perf-changelog.yaml | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/perf-changelog.yaml b/perf-changelog.yaml index b6b8f6499..1a63fb1d1 100644 --- a/perf-changelog.yaml +++ b/perf-changelog.yaml @@ -1282,6 +1282,5 @@ - config-keys: - minimaxm2.5-fp8-b200-vllm description: - - "Update MiniMax-M2.5 FP8 B200 config with new search spaces and vllm args" + - "Update MiniMax-M2.5 FP8 B200 config with new search spaces" pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1010 - \ No newline at end of file