2 ms·
No, for now, I’ve only made it work with Ollama, but it could be ideal to do it directly on llama.cpp. Thank you, I’ll take note of it.
by Dontizi 2y ago
No, for now, I’ve only made it work with Ollama, but it could be ideal to do it directly on llama.cpp. Thank you, I’ll take note of it.
- car 2y agoThat would be great. Llama.cpp’s built in server offers HTTP embedding endpoints.