Recursive releases Flint, a free coding model built to run on your own machine

Recursive, the AI startup led by Richard Socher, has released Flint, an open-weight coding model on Hugging Face under Apache 2.0. It is a mixture-of-experts model with 40 billion parameters in total, of which only 2.5 billion are active for each token, which keeps it fast on local hardware. Recursive says it produces more than 125 tokens a second on an M4 Pro MacBook Pro and more than 150 on a 64GB NVIDIA RTX Spark, and that it leads mixture-of-experts models under 50 billion parameters on agentic coding benchmarks; these are the company's own tests. It works with llama.cpp, Ollama, LM Studio, Unsloth, vLLM and SGLang. Recursive says its automated research system, the Eureka Machine, helped build nearly every part of the model, and that a small team built it from scratch in under four months.