Press Space to continue
Local 40B coding model runs 125+ tokens per second on MacBook
Press Space to continue
wtf this is actually insane: > new 40B AI coding model called Flint > runs locally on your MacBook or RTX hardware > 125+ tokens/sec on MacBook at short contexts > open weights under Apache 2.0 > no API bills, no token limits > built from scratch in under 4 months > AI helped design and improve the AI itself crazy what @recursive_si are doing: AI builds better AI
RichardSocher@RichardSocher3d ago
Introducing Flint, @Recursive_SI’s first model: a fast local coding model built with our automated research system, the Eureka Machine. • Best-in-class fast local coding • Up to 2x faster than comparable models: 125+ tokens/sec on a MacBook Pro, 150+ on RTX Spark • Built from scratch in under 4 months Private, unmetered intelligence on your own machine. Open weights available now on @huggingface, run it locally on your favorite framework or try it for free on @opencode or @pidotdev.
