Skip to content

[BUG]: custom embedding model error #351

@Jays-1111

Description

@Jays-1111

🐛 Bug description [Please make everyone to understand it]

因为当前0.1.8无法使用自定义模型,所以使用0.1.7配置自定义模型api
现象:使用本地flm server(https://fastflowlm.com/)作为自定义模型API,设置vlm模型和embedding模型,embedding模型出现错误无法连接
模型为:embed-gemma:300m
Image

🧑‍💻 Step to reproduce

1.download flm server: https://fastflowlm.com/
2.flm serve --ctx-len 32768 -s 32 -q 32
3.Minecontext 0.1.7:URL:http://127.0.0.1:52625/v1 model:embed-gemma:300m URL:http://127.0.0.1:52625/v1 model:qwen3vl-it:4b
4.save-出现错误

👾 Expected result

save 之后显示连接成功
可以成功使用本地embedding 模型

🚑 Any additional information

No response

🛠️ MineContext Version

0.1.7

💻 Platform Details

windows11

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions