LLMUnity

LLMUnity copied to clipboard

[Android build] IOException: The file exists

1

### Describe the bug When I build the demo android sample, It's successfully at first time, but I rebuild, it will show the error ### Steps to reproduce 1. Build...

siduko

Issue starting LLM

5

### Describe the bug LLM fails to start up on a build on a laptop with these specs: Processor: AMD Ryzen 5 7520U with Radeon Graphics Video Card: AMD Radeon(TM)...

Eelguezabal

amakropoulos

Unable to load Hubble-4B

1

### Describe the bug Model URL: https://huggingface.co/bartowski/Hubble-4B-v1-GGUF/discussions/1 llama_model_loader: - kv 26: tokenizer.ggml.merges arr[str,280147] = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "... llama_model_loader: - kv 27: tokenizer.ggml.bos_token_id u32 = 128000 llama_model_loader:...

tempstudio

### Describe the feature Now that Embedding functionality is implemented directly in LLMUnity (v2.2.0), the RAG functionality from [RAGSearchUnity](https://github.com/undreamai/RAGSearchUnity) can be migrated directly in LLMUnity.

amakropoulos

mmproj model loading and image feature extraction update. You will need to load a vision model and its mmproj file. The settings are in the "LLM.cs" script under the "Advanced...

NOSCOPEdev

added UseNativeDebugger which will pause the loading of LLM to help attaching c++ debugger. it will discard release dlls. ideally there will be just one dll compiled with debug enabled

psocolovsky

amakropoulos

amakropoulos

amakropoulos