Recursive SI Releases Flint: Open-Weights Local Coding Model Hitting 125+ tokens/sec

ariG23498 · x · 2026-10-22

• Recursive SI, Richard Socher's startup, launched Flint, its first model targeting fast local coding, built from scratch in under 4 months by its automated research system, the Eureka Machine.
• Claims up to 2x faster than comparable models: 125+ tokens/sec on a MacBook Pro and 150+ on RTX Spark, pitched as best-in-class fast local coding.
• Open weights are live on Hugging Face, offering private, unmetered on-device intelligence; the chat template supports reasoning traces and tool calling.

Related event: Recursive Open-Sources Flint, a 40B Local Coding MoE Built by Its Automated Eureka Research System in Under 4 Months(101 posts)→

Original post →

More from coding & agent

• Swarm-lite experiment lets worker agents broadcast and talk to each other in SENPAI — morgymcg · 2026-10-22
• Solo dev splits MacBook, Mac mini and a server to run his posting agents — petrusenko_max · 2026-10-22
• Anthropic safety guardrails block LLM-driven crypto trading agent, dev reports — devanshmehta · 2026-10-22
• AI agent fixes real CSV bug in Docker sandbox, pushes branch and opens PR in ~45s — Acceptable-Object390 · 2026-10-22
• mdancho84: The Valuable Ones Architect Systems — Agents, Workflows, Memory, Eval Loops — mdancho84 · 2026-10-22
• AutoResearch: AI Agents That Run the Full Research Loop Autonomously — mdancho84 · 2026-10-22

coding & agent channel →