Support Text-Generation Webui

[iChristGit]

⚠️ Please check that this feature request hasn't been suggested before.

• [X] I searched previous Ideas in Discussions didn't find any similar feature requests.
• [X] I searched previous Issues didn't find any similar feature requests.

🔖 Feature description

This is the project: https://github.com/oobabooga/text-generation-webui/ It has by default an openai like api and it all runs locally, can it be support with this project?

✔️ Solution

Add support for TextGenWebui

❓ Alternatives

None

📝 Additional Context

No response

Acknowledgements

• [X] My issue title is concise, descriptive, and in title casing.
• [X] I have searched the existing issues to make sure this feature has not been requested yet.
• [X] I have provided enough information for the maintainers to understand and evaluate this request.

[SubGlitch1]

you can. in the .env just change the base url api

[SubGlitch1]

u have to launch text-generation with the openai extention. the base url then becomes https://127.0.0.1:5000

[the-cat-crying]

Tools Select what external tools your agents have access to.

Error loading tools.

Language 简体中文

API Key You can optionally use your own API key here. You can find your API key in your OpenAI dashboard.

http://localhost:8051/v1

Test Advanced Settings Model gpt-3.5-turbo

and http://localhost:8051/v1 is local llm api,but it not work.

[yhyu13]

Ehhh, textgen webui does not support function calling at this moment, so the api call would fail immediately

[chymian]

we getting closer for ooba's function calling support: PR 5185 is nearly ready to be merged.

[zwilch]

cd next/src grep -R "openai"

there is openai.com hardcoded and not using .env

server/api/routers/agentRouter.ts the functionnew OpenAI() has no baseURL:

[skywalker123p]

Harcoded OpenAPI url should be replace by configurable so we can use any custom locally running LLM.