Recursive Opens Flint, a 40B/2.5B MoE Coder Hitting 89.7% on SWE-bench Pro v2 Under Apache 2.0

Recursive Superintelligence's first model, Flint, is a 39.8B-parameter MoE with 2.5B active, 256 routed experts (6 selected per token), 262K context, Apache 2.0 weights, and quantizations down to 11.4 GB. Benchmarks (self-reported): 89.7% SWE-bench Pro v2, 74.2% SWE-bench Verified, 54.9% Terminal-Bench 2.1, with ~125 tokens/sec on an M4 Pro MacBook. Recursive says most of the development was automated by its 'Eureka Machine' self-improvement system.