Recursive team on shipping Flint: full pre-training to RL pipeline built from scratch

ChengleiSi · x · 2026-10-22

Recursive AI team members shared the launch of Flint, the startup's first model: a fast local coding model whose entire pipeline — pre-training, mid-training, and reinforcement learning — was built from scratch by a small team with help from their automated research system. It runs at 125+ tokens/sec on a MacBook Pro (150+ on RTX Spark), was trained in under 4 months, and has open weights on Hugging Face.

Related event: Recursive releases Flint, an open-weights 40B MoE coding model built in under 4 months by its automated Eureka research system(50 posts)→

Original post →

More from Models

• Recursive AI ships Flint, a 40B-A2.5B MoE claimed to match Qwen3.6-27B, weights on Hugging Face — ChengleiSi · 2026-10-22
• DeepSeek ships V4.1-Flash-1017, now the #1 open model on RSI-Inference Bench — scaling01 · 2026-10-22
• OpenAI's Cybersecurity Guardrails Block a Dev From Auditing His Own Patched Server — Next_Technology6361 · 2026-10-22
• TPN-006 hits Hugging Face: 7.7GB compressed model beats its 17GB Mistral teacher — markjeffrey · 2026-10-22
• Kimi K3.1's latest checkpoint draws 49k-token SVG portrait zero-shot in 22 minutes — lyraxana · 2026-10-22
• MiniMax launches M3.1 Flash at 1/3 the cost and 3x the speed of M3, weights out Oct 23 — MiniMax_AI · 2026-10-22

Models channel →