Closed artivis closed 3 months ago
Thanks for opening this issue! This has now been implemented and released (run hishtory update
to get the latest version). You can configure a custom AI API endpoint via hishtory config-set ai-completion-endpoint https://your-openai-compatible-endpoint.example/path
. If you run into any issues with this, please let me know
Hello,
Would it be possible to extend the AI support? Especially to offer the possibility to hit self-hosted instances. I guess the first low hanging fruit would be to make the OpenAI endpoint and model parameterizable. This would allow to hit OpenAI compatible projects such as ollama.
Thanks.