{"schema":"postcutoff/model@1","as_of":"2026-10-22T13:31:00+02:00","url":"https://postcutoff.com/m/recursive-flint/","md":"https://postcutoff.com/m/recursive-flint/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"recursive-flint","name":"Recursive Flint","org":"Recursive","family":"Flint","released":"2026-10-21","status":"current","type":"code","modality_in":["text"],"modality_out":["text"],"open_weights":true,"model_license":"apache-2.0","context_window":262144,"max_output":null,"knowledge_cutoff":"2026-06","pricing":null,"price_line":"Price not published","access":[{"provider":"Hugging Face","url":"https://huggingface.co/RecursiveSI/flint"},{"provider":"Hugging Face (int4 / NVFP4 GGUF)","url":"https://huggingface.co/RecursiveSI/flint-int4-gguf"},{"provider":"Recursive local engine","url":"https://github.com/recursive-oss/rec","docs":"https://docs.recursive.com/getting-started"}],"capabilities":[{"name":"Model largely developed by automated AI research","detail":"Recursive says its Eureka Machine helped develop nearly every part of Flint's model-building stack, from data and training recipes to infrastructure and inference optimization.","first":false,"discovered":"launch","source":"https://www.recursive.com/updates/introducing-recursive-flint"},{"name":"Fast local agentic coding","detail":"40B-total / ~2.5B-active MoE; 125+ tokens/s on an M4 Pro MacBook Pro and 150+ on RTX Spark; 74.2% SWE-bench Verified, 54.9% Terminal-Bench 2.1 (developer-reported).","first":false,"discovered":"launch","source":"https://huggingface.co/RecursiveSI/flint"}],"entry":"2026-10-21-recursive-flint-eureka-machine","notes":"Open-weight coding MoE (39.8B total, ~2.5B active, 262k context). BF16 checkpoint ~80 GB serves on one B300/H200 via the bundled SGLang kit; Transformers modeling code not included at launch. Recursive recommends splitting big tasks into smaller steps. Also free on OpenCode and pi.dev (per Socher).","verified":"2026-10-22","body_md":"Quick start (model card): `hf download RecursiveSI/flint`, build the SGLang image in `sglang/`, then call the OpenAI-compatible endpoint at\n`http://localhost:30000/v1` with `model=\"flint\"`.","page_url":"https://postcutoff.com/m/recursive-flint/","standing":{"kind":"code","newer_count":0,"newer":[],"before":null,"after":null,"weights":"public","licence":"apache-2.0","licence_permissive":true,"summary":"Weights are public under a permissive licence (Apache-2.0)."}}