From 6c1bb27b6916fb6450b1ec1ce9fef81474e15881 Mon Sep 17 00:00:00 2001 From: akshatvishu <33392262+akshatvishu@users.noreply.github.com> Date: Sun, 24 May 2026 22:21:39 +0530 Subject: [PATCH] docs: fix CUDA pre-built image command (#3836) Signed-off-by: akshatvishu --- docs/getting_started/installation/gpu/cuda.inc.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/getting_started/installation/gpu/cuda.inc.md b/docs/getting_started/installation/gpu/cuda.inc.md index 69fc05351..9f3c86843 100644 --- a/docs/getting_started/installation/gpu/cuda.inc.md +++ b/docs/getting_started/installation/gpu/cuda.inc.md @@ -103,10 +103,10 @@ docker run --runtime nvidia --gpus 2 \ -p 8091:8091 \ --ipc=host \ vllm/vllm-omni:v0.21.0 \ - --model Qwen/Qwen3-Omni-30B-A3B-Instruct --port 8091 + vllm serve Qwen/Qwen3-Omni-30B-A3B-Instruct --omni --port 8091 ``` !!! tip - You can use this docker image to serve models the same way you would with in vLLM! To do so, make sure you overwrite the default entrypoint (`vllm serve --omni`) which works only for models supported in the vLLM-Omni project. + The CUDA image does not define a default entrypoint, so include `vllm serve ... --omni` after the image name. # --8<-- [end:pre-built-images]