LLM
A blazing fast inference solution for text embeddings models
Rust client for the huggingface hub aiming for minimal subset of features over `huggingface-hub` python package
💥 Fast State-of-the-Art Tokenizers optimized for Research and Production
High-performance runtime for data analytics applications
Codes/Notebooks for AI Projects
Burn is a next generation tensor library and Deep Learning Framework that doesn't compromise on flexibility, efficiency and portability.
A general memory system for agents, powered by deep-research
TensorZero is an open-source stack for industrial-grade LLM applications. It unifies an LLM gateway, observability, optimization, evaluation, and experimentation.
Privacy first, AI meeting assistant with 4x faster Parakeet/Whisper live transcription, speaker diarization, and Ollama summarization built on Rust. 100% local processing. no cloud required. Meetil…

