Edisonwei54

> 是的，这将在以后修复，您可以从[这里](https://github.com/eosphoros-ai/DB-GPT-Web/issues/74)找到 请问修复了吗？我使用最新版本，还是遇到了同样的问题

> API更新 [4afecd1](https://github.com/RVC-Boss/GPT-SoVITS/commit/4afecd1950845a974350cd2d8dc8dcf12398fba9) 我现在测试api的流式输出，感觉好像还是模型推理结束之后，把数据用流式的形式返回，而不是一边推理一边返回吧

> 切割使用，分段流返 但是在测试的时候，效果还是切割的片段全部推理结束之后才会流式分段返回，确实在播放的时候看到了时间上的自增

@WoosukKwon How can I solve this problem

> Current punica kernel can't process ` h_out=3424` , you can set `-tensor-parallel-size 2` to avoid this error Thanks, It can work now, but I still want to use all...

Traceback (most recent call last): File "/opt/conda/envs/vllm/lib/python3.10/site-packages/vllm/lora/worker_manager.py", line 150, in _load_lora lora = self._lora_model_cls.from_local_checkpoint( File "/opt/conda/envs/vllm/lib/python3.10/site-packages/vllm/lora/models.py", line 246, in from_local_checkpoint return cls.from_lora_tensors( File "/opt/conda/envs/vllm/lib/python3.10/site-packages/vllm/lora/models.py", line 150, in from_lora_tensors module_name, is_lora_a...

@WoosukKwon @zhuohan123 https://github.com/vllm-project/vllm/pull/3177 I see that this submission already supports Lora for Qwen2. What is the reason for it still not working? Is it due to Lora's issue?

同样的问题，有什么解决的方法吗

同样的问题，有什么解决办法吗

@Jintao-Huang @tastelikefeet