From 3b00bcd0f0349d7c4767790746122bb63e18d4c8 Mon Sep 17 00:00:00 2001 From: functionstackx <47992694+functionstackx@users.noreply.github.com> Date: Sat, 4 Jul 2026 17:19:24 -0400 Subject: [PATCH] [Klaud Cold] Update qwen3.5-fp4-b200-sglang (+mtp) SGLang image to v0.5.14-cu130 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 中文:将 qwen3.5-fp4-b200-sglang(+mtp) 的 SGLang 镜像 升级至 v0.5.14-cu130。 Co-Authored-By: Claude Fable 5 --- configs/nvidia-master.yaml | 4 ++-- perf-changelog.yaml | 7 +++++++ 2 files changed, 9 insertions(+), 2 deletions(-) diff --git a/configs/nvidia-master.yaml b/configs/nvidia-master.yaml index bdb6a46029..f8e97b565f 100644 --- a/configs/nvidia-master.yaml +++ b/configs/nvidia-master.yaml @@ -2125,7 +2125,7 @@ qwen3.5-fp8-b200-sglang-agentic: qwen3.5-fp4-b200-sglang: - image: lmsysorg/sglang:v0.5.12-cu130 + image: lmsysorg/sglang:v0.5.14-cu130 model: nvidia/Qwen3.5-397B-A17B-NVFP4 model-prefix: qwen3.5 runner: b200 @@ -2146,7 +2146,7 @@ qwen3.5-fp4-b200-sglang: - { tp: 2, ep: 1, conc-start: 4, conc-end: 128 } qwen3.5-fp4-b200-sglang-mtp: - image: lmsysorg/sglang:v0.5.12-cu130 + image: lmsysorg/sglang:v0.5.14-cu130 model: nvidia/Qwen3.5-397B-A17B-NVFP4 model-prefix: qwen3.5 runner: b200 diff --git a/perf-changelog.yaml b/perf-changelog.yaml index f13d283d65..9f50a49f0f 100644 --- a/perf-changelog.yaml +++ b/perf-changelog.yaml @@ -4448,3 +4448,10 @@ - "Employ the AITER MLA attention backend for the DeepSeek-V4 MLA path." - "Switch the MoE backend from triton_unfused to AITER MoE (VLLM_ROCM_USE_AITER_MOE=1 + --moe-backend aiter) for the FP4 experts." pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1980 + +- config-keys: + - qwen3.5-fp4-b200-sglang + - qwen3.5-fp4-b200-sglang-mtp + description: + - "Update SGLang image from lmsysorg/sglang:v0.5.12-cu130 to lmsysorg/sglang:v0.5.14-cu130" + pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2056