From b3ead32d3ca276d3d0d9ab4169c4700ddf047161 Mon Sep 17 00:00:00 2001 From: Baizhou Zhang Date: Wed, 29 Apr 2026 13:26:13 -0700 Subject: [PATCH] [minor] Remove incorrect note after supporting w4a16 moe for DeepSeek V4 (#24035) --- docs_new/cookbook/autoregressive/DeepSeek/DeepSeek-V4.mdx | 3 --- 1 file changed, 3 deletions(-) diff --git a/docs_new/cookbook/autoregressive/DeepSeek/DeepSeek-V4.mdx b/docs_new/cookbook/autoregressive/DeepSeek/DeepSeek-V4.mdx index 86f5778d3..0e01e0a35 100644 --- a/docs_new/cookbook/autoregressive/DeepSeek/DeepSeek-V4.mdx +++ b/docs_new/cookbook/autoregressive/DeepSeek/DeepSeek-V4.mdx @@ -120,9 +120,6 @@ docker run --gpus all \ SGLang supports three main serving recipes for DeepSeek-V4 with different latency/throughput trade-offs (`low-latency`, `balanced`, `max-throughput`), plus specialized recipes for long-context (`cp`, prefill context-parallel) and prefill/decode disaggregation (`pd-disagg`). The interactive generator below emits the exact launch command for any `(hardware, variant, recipe)` combination. - -For H200 GPU deployments, use the SGLang checkpoint under `sgl-project`, not the default DeepSeek checkpoint. - ### 3.1 Basic Configuration