Recursive_SI launches Flint: open-weight local coding model hits 125+ tokens/sec on MacBook
HeyAmit_ · x · 2026-10-22
RecursiveSI, founded by Richard Socher, has released Flint, its first model — a fast local coding model built from scratch in under 4 months by the company's automated research system, the Eureka Machine.
- Performance: Claims best-in-class fast local coding, up to 2x faster than comparable models — 125+ tokens/sec on a MacBook Pro, 150+ on RTX Spark.
- Open weights: Available now on Hugging Face; runnable on any local framework, with free trials via opencode and pidot.dev.
- Pitch: Private, unmetered intelligence on your own machine, emphasizing control over code and workflow.
The post was amplified by HeyAmit, who framed it as why local AI matters: coding assistants without a cloud API behind every prompt.
More from coding & agent
- Beacon open-sources a cross-harness self-improving memory layer for 20+ coding agents — solyarisoftware · 2026-10-22
- After 4 CVSS-10 CVEs in 24 hours, a team builds an auto-SRE tool that fixes itself — viksit · 2026-10-22
- Pi v1.4.0 ships image classification in codemode and fixes summary truncation — solyarisoftware · 2026-10-22
- Claude Code 2.1.308 Adds ~11K Prompt Tokens Just 8 Hours After 2.1.307 — ClaudeCodeLog · 2026-10-22
- Claude Code 2.1.308 ships CLI bug fixes and reliability improvements — ClaudeCodeLog · 2026-10-22
- OpenAI Engineer Asks Users What Would Make the OpenAI CLI More Useful — pbbakkum · 2026-10-22