Andrei
Andrei
Weights and example code: https://huggingface.co/abetlen/paligemma-3b-mix-224-gguf - the chat handler here only supports single image / single turn conversations, just did this to simplify testing but will change before merging -...
- [x] Use `llama_decode` instead of deprecated `llama_eval` in `Llama` class - [ ] Implement batched inference support for `generate` and `create_completion` methods in `Llama` class - [ ] Add...