Llama is an open-source toolkit for training and fine-tuning large language models (LLMs). It provides tools for efficient model development, including data preprocessing, training scripts, and model evaluation. Suitable for research and production, Llama supports various architectures and scales to accommodate different hardware setups.
This project creates a real-time conversational AI, either serverless via SvelteKit/Static or using LangChain with FastAPI as a web server, streaming GPT model responses and supporting in-browser LLMs via webllm
CSVs of the Huggingface and LMSYS LLM leaderboards, along with the code to generate them in R