launch audio model-cosyvoice-300M-sft ERROR

[DoItEric]

System Info / 系統信息

env: docker 27.2.0 // ubuntu 22.04 LTS image: xprobe/xinference:v0.15.0

Running Xinference with Docker? / 是否使用 Docker 运行 Xinfernece？

• [X] docker / docker
• [ ] pip install / 通过 pip install 安装
• [ ] installation from source / 从源码安装

Version info / 版本信息

env: docker 27.2.0 // ubuntu 22.04 LTS image: xprobe/xinference:v0.15.0 devices: NVIDIA 3090 * 2

The command used to start Xinference / 用以启动 xinference 的命令

docker run --name "$CONTAINER_NAME" -d \
    -p 9997:9997 \
    -e XINFERENCE_HOME=/data \
    -v /root/apps/xinference/data:/data \
    -e HF_ENDPOINT="https://hf-mirror.com" \
    --gpus all \
    xprobe/xinference:v0.15.0 \
    xinference-local -H 0.0.0.0

Reproduction / 复现过程

1. run docker image
2. launch cosyvoice-300M-sft
3. get error : Server error: 500 - [address=0.0.0.0:41795, pid=968] ('Invalid argument to class cosyvoice.llm.llm.TransformerLM', "TransformerLM.init() missing 1 required positional argument: 'sampling'")

looks like code error? i guess is code version conflict？

Expected behavior / 期待表现

just no bug

[leslie2046]

you need to update the models from huggingface/modelscope which is updated last week

[github-actions[bot]]

This issue is stale because it has been open for 7 days with no activity.

[github-actions[bot]]

This issue was closed because it has been inactive for 5 days since being marked as stale.