Is it still intended and possible to use custom endpoint in 0.2.0?

[Kaschi14]

No description provided.

[senbinyu]

same question. I can use my own custom endpoint in the previous version, but it can not work in the newest 0.2.0 version with the same configuration.

[github-actions]

This issue is stale because it has been open for 30 days with no activity.

[f-ricci]

I have the same question

[github-actions]

This issue is stale because it has been open for 30 days with no activity.

[McPatate]

Sorry for the late response. What do you mean by custom endpoint? Could you share your configuration? llm-vscode supports multiple backends, namely Hugging Face, TGI, OpenAI, ollama & llama.cpp

[senbinyu]

Sorry for the late response. What do you mean by custom endpoint? Could you share your configuration? llm-vscode supports multiple backends, namely Hugging Face, TGI, OpenAI, ollama & llama.cpp

@McPatate In the previous version of llm-vscode, I can use my own local deployed model. For example, I deployed my model at the port: http:https://192.168.1.73:8192/generate, then I can put the address at ModelID or Endpoint to obtain the response. But in the newest version of 0.2.0, it does not work anymore. Great thanks.

[McPatate]

Could you share your configuration settings, any logs or anything else that could help understand what is going on?

Have you tried updating to the latest version of llm-vscode?

[github-actions]

This issue is stale because it has been open for 30 days with no activity.

[McPatate]

I'll close this issue, feel free to re-open if you are still facing an issue.