Flint is an open-weight AI model that runs on local hardware
Its developer says Flint exceeds 125 tokens per second on an M4 Pro MacBook Pro at short contexts.
TLDR
Flint’s developer says the model runs on users’ own hardware without API fees. They say a small team built it from scratch in under four months with an automated research system, and claim it leads mixture-of-experts models under 50 billion total parameters on agentic coding benchmarks.
Combined views
820.2K
102 Sources, first seen ago

