How to set detailed configuration when using lmdeploy?

Open davidluciolu opened this issue 9 months ago • 1 comments

For example, I want to test the Qwen2-VL-2B-Instruct model with some configurations like max_pixels= 6272000. How to set this when using lmdeploy?

What's more, how to use custom prompt when using lmdeploy?

Can you provide the example command?

@PhoenixZ810

Mar 21 '25 04:03 davidluciolu

Hello @davidluciolu , this setting can be configured for the incoming text/images when passing messages, please refer to to https://github.com/InternLM/lmdeploy/blob/main/docs/en/multi_modal/qwen2_5_vl.md#offline-inference for detailed solution.

Mar 27 '25 12:03 FangXinyu-0913