目前OpenAIServer是不是只支持LLM不支持VLM？

[llery]

使用TensorRT-LLM v0.15.0自带的OpenAIServer来启动一个VLM视觉模型后，发现messages只能传type=text类型的content，如果需要附带图片信息image或image_url是不支持的（等于是只能推理LLM），想问一下这个问题是暂时的吗？后期有计划会解决吗？（源码中runtime/multimodal_model_runner.py中是支持常见的VLM视觉模型推理的，暂时就只能用multimodal_model_runner吗？？）

[nv-guomingz]

@LinPoly would u please take a look this question

[LinPoly]

As far as I know, VLMs are not yet supported in our LLM API and OpenAIServer is built atop LLM API, so we don't have VLMs in our OpenAIServer, we will supported them when LLM API is ready.

[nv-guomingz]

Hi @llery Please feel free to reopen this ticket if needed