codingl2k1
codingl2k1
CosyVoice几小时前还在推代码，等稍微稳定些再同步更新。
> [@codingl2k1](https://github.com/codingl2k1) Could you look at this? 我看看
The tool call for the vLLM backend is currently not supported. You can try using the transformers backend to see if it supports tool calls.
服务的还有错误日志吗？
有尝试使用 xinference client 吗？如果要自己用 requests 实现请参考： https://github.com/xorbitsai/inference/blob/main/xinference/client/restful/restful_client.py#L764
> 同问 有没有 xinference 服务端日志啊？
Same issue, I am using the CPU build: CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 |...
> Re-download Model. Try it. work for me. > > ``` > # SDK模型下载 > from modelscope import snapshot_download > snapshot_download('iic/CosyVoice-300M', local_dir='pretrained_models/CosyVoice-300M') > snapshot_download('iic/CosyVoice-300M-SFT', local_dir='pretrained_models/CosyVoice-300M-SFT') > snapshot_download('iic/CosyVoice-300M-Instruct', local_dir='pretrained_models/CosyVoice-300M-Instruct') > snapshot_download('iic/CosyVoice-ttsfrd',...
The models on huggingface are outdated. 
Lingering Ghost Processes with 0MiB GPU Memory Reported After Model Actor Creation Observed Behavior
Could you please try the latest version?