Support integrated local llm runtimes like llama.cpp. This means user can easily download models and use them directly without need to have Ollama installed or other software.
https://github.com/ggml-org/llama.cpp
Loading models directly from Hugging Face listings?
User Interface
There needs to be multiple settings screens for:
- managing local runtime
- managing and downloading models
- managing storage ?
Alternatives?
Another option would be to somehow integrate Ollama into this app?
This would be for discussion.
Support integrated local llm runtimes like llama.cpp. This means user can easily download models and use them directly without need to have Ollama installed or other software.
https://github.com/ggml-org/llama.cpp
Loading models directly from Hugging Face listings?
User Interface
There needs to be multiple settings screens for:
Alternatives?
Another option would be to somehow integrate Ollama into this app?
This would be for discussion.