From 3dc9ff3ce8bb88dcbcf2655f616bd5439f224c11 Mon Sep 17 00:00:00 2001 From: Shenggui Li Date: Wed, 26 Feb 2025 11:40:47 +0800 Subject: [PATCH] [doc] fixed dpsk quant faq (#3865) --- docs/references/deepseek.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/references/deepseek.md b/docs/references/deepseek.md index 883dd6119..ff3d649de 100644 --- a/docs/references/deepseek.md +++ b/docs/references/deepseek.md @@ -90,8 +90,8 @@ Overall, with these optimizations, we have achieved up to a 7x acceleration in o 2. **Question**: How to use quantized DeepSeek models? - **Answer**: DeepSeek's MLA does not have support for quantization. You need to add the `--disable-mla` flag to run the quantized model successfully. Meanwhile, AWQ does not support BF16, so add the `--dtype half` flag if AWQ is used for quantization. One example is as follows: + **Answer**: AWQ does not support BF16, so add the `--dtype half` flag if AWQ is used for quantization. One example is as follows: ```bash - python3 -m sglang.launch_server --model cognitivecomputations/DeepSeek-R1-AWQ --tp 8 --trust-remote-code --dtype half --disable-mla + python3 -m sglang.launch_server --model cognitivecomputations/DeepSeek-R1-AWQ --tp 8 --trust-remote-code --dtype half ```