[Feature] Hope to support the image understanding model deployed through xinference

[dlluckboy]

MaxKB Version

1.9.1

Please describe your needs or suggestions for improvements

通过xinference部署的deepseek-vl-chat，glm-4v，qwen2-vl模型，无法在maxkb中作为图片理解模型进行添加，会提示报错。

Please describe the solution you suggest

No response

Additional Information

No response

[zyyfit]

感谢反馈，后续版本修复

[shaohuzhang1]

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

Thanks for the feedback, it will be fixed in subsequent versions.

[baixin513]

目前排查下了应该是所选模型的部分参数无法兼容，您可以换个模型试一下，暂时无法解决

[shaohuzhang1]

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

Currently, the check has been made that some parameters of the selected model are not compatible. You can try it with a different model, but it cannot be solved for the time being.