From 700f2ee4c53df34968c41ce6f2cc99c3d7c76b26 Mon Sep 17 00:00:00 2001 From: "claude[bot]" <41898282+claude[bot]@users.noreply.github.com> Date: Fri, 23 Jan 2026 00:28:07 +0000 Subject: [PATCH 1/2] Update dsr1-fp8-h200-sglang to SGLang v0.5.7 Update H200 DeepSeek R1 FP8 SGLang image from v0.5.6 to v0.5.7. Closes #389 Co-authored-by: functionstackx --- .github/configs/nvidia-master.yaml | 2 +- perf-changelog.yaml | 6 ++++++ 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/.github/configs/nvidia-master.yaml b/.github/configs/nvidia-master.yaml index 06c37888a..e2900490d 100644 --- a/.github/configs/nvidia-master.yaml +++ b/.github/configs/nvidia-master.yaml @@ -205,7 +205,7 @@ dsr1-fp8-b200-trt-mtp: - { tp: 8, ep: 8, dp-attn: true, conc-start: 64, conc-end: 256, spec-decoding: mtp } dsr1-fp8-h200-sglang: - image: lmsysorg/sglang:v0.5.6-cu129-amd64 + image: lmsysorg/sglang:v0.5.7-cu129-amd64 model: deepseek-ai/DeepSeek-R1-0528 model-prefix: dsr1 runner: h200 diff --git a/perf-changelog.yaml b/perf-changelog.yaml index 71a44dabb..1b80f5c7d 100644 --- a/perf-changelog.yaml +++ b/perf-changelog.yaml @@ -195,3 +195,9 @@ - "Remove deprecated --max-seq-len-to-capture flag" - "Add HIP_VISIBLE_DEVICES env var for Ray compatibility in vLLM 0.14+" pr-link: https://github.com/InferenceMAX/InferenceMAX/pull/496 + +- config-keys: + - dsr1-fp8-h200-sglang + description: + - "Update H200 DeepSeek R1 FP8 SGLang image from v0.5.6 to v0.5.7" + pr-link: https://github.com/InferenceMAX/InferenceMAX/pull/XXX From 4c94823bf46076fdac8f66b5702d62dcaa8a5cba Mon Sep 17 00:00:00 2001 From: functionstackx <47992694+functionstackx@users.noreply.github.com> Date: Thu, 22 Jan 2026 22:25:37 -0500 Subject: [PATCH 2/2] Update perf-changelog.yaml Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com> --- perf-changelog.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/perf-changelog.yaml b/perf-changelog.yaml index 1b80f5c7d..087184f08 100644 --- a/perf-changelog.yaml +++ b/perf-changelog.yaml @@ -200,4 +200,4 @@ - dsr1-fp8-h200-sglang description: - "Update H200 DeepSeek R1 FP8 SGLang image from v0.5.6 to v0.5.7" - pr-link: https://github.com/InferenceMAX/InferenceMAX/pull/XXX + pr-link: https://github.com/InferenceMAX/InferenceMAX/pull/538