verigle
verigle
> ```python > import requests > > class DifyClient: > def __init__(self, api_key): > self.api_key = api_key > self.base_url = "https://api.dify.ai/v1" > > def _send_request(self, method, endpoint, json=None, params=None, stream=False):...
is there any plan to release docker image for ubuntu user ?
this PR is not usefull for me, I want know how to transfor the [docs of vllm-on-multiple-nodes](https://docs.vllm.ai/en/latest/serving/distributed_serving.html#running-vllm-on-multiple-nodes) to the docker-compose.yml file
sorry, if vllm still version 0.6.0, fail to install ffmpeg, because the base image of vllm 0.6.0 is ubuntu 20.04, test is not ok.
I expect the tts don't read the markdown block of image url. example for llm output: ``` this is a new picture:  ``` for tts, it only required to...
is there any plan to filter the markdown image block for TTS input?
多模态模型是否可以支持显存均匀分布到多卡
更新的哪个版本的镜像？， docker-compose.yml 如何修改的？
is there has any plan to release training code of google/timesfm-1.0-200m-pytorch?