GitHub

Run any AI model locally — one unified API for chat, vision, speech, images, video, and more. Built for multi-user serving.

InferAll is a self-hosted inference server that exposes an OpenAI-compatible REST API for every type of AI model. Point any OpenAI SDK client, LangChain, LlamaIndex, or custom application at InferAll ...
LinkedIn

A Quick Introduction to Recursive Language Models (RLMs)

In the rapidly evolving landscape of Large Language Models (LLMs), the industry has been obsessed with a single metric: the context window. We’ve moved from 8k to 128k, and now to 2 million tokens.
VentureBeat

MIT’s new ‘recursive’ framework lets LLMs process 10 million tokens without context rot

Recursive language models (RLMs) are an inference technique developed by researchers at MIT CSAIL that treat long prompts as an external environment to the model. Instead of forcing the entire prompt ...
Nature

A comprehensive foundation model for cryo-EM image processing

Cryogenic electron microscopy (cryo-EM) has become a premier technique for determining high-resolution structures of biological macromolecules. However, its broad application is constrained by the ...
LinkedIn

NuminaMath 7B TIR: A New Era in AI-Powered Mathematical Problem-Solving

Numina has announced the release of its latest model, NuminaMath 7B TIR, a groundbreaking advancement in the realm of mathematical problem-solving. This sophisticated language model, equipped with ...
GitHub

Rigel is a language for describing image processing hardware embedded in Lua. Rigel can compile to Verilog hardware designs for Xilinx FPGAs, and also can compile to fast x86 ...

Installing Rigel is not required. Rigel requires luajit, which can be easily installed with either apt-get or homebrew. Rigel provides a wrapper script (rigelLuajit ...