Recursive SI Releases Flint, a 40B MoE Local Coding Model Hitting 125+ tok/s on MacBook
_rockt · x · 2026-10-22
Recursive SI has released Flint, its first model: a 40B MoE built for fast, local agentic coding, with open weights now on Hugging Face.
- Performance: leads its class on SWE-Bench Pro v2 and Terminal-Bench 2.1; up to 2x faster than comparable models at 125+ tokens/sec on a MacBook Pro and 150+ on RTX Spark.
- How it was built: Recursive's automated research system, the Eureka Machine, designed its sparse shape, low-bit training and kernels; built from scratch in under 4 months.
- Usage: runs fully local and offline so code stays on-device; SGLang offers Day-0 support, a DGX Spark cookbook recipe is available, and it can be tried free on opencode or pidotdev.
More from coding & agent
- Migrating an entire Hermes automation workload to Antigravity took about 2 hours — rseroter · 2026-10-22
- How a second Mac mini lets AI agents auto-distribute compute-heavy tasks — tedddyoweh · 2026-10-22
- LangChain's Design Agent Shipped 3,000+ Assets and Earned 4.5M Views Since August — LangChain · 2026-10-22
- Databricks Chief AI Scientist: Evals Are the Biggest Bottleneck to Enterprise AI Adoption — jefrankle · 2026-10-22
- Developer plans a solar-aware job scheduler so his homelab rides free excess power — mertdumenci · 2026-10-22
- How do you design your agentic code review workflow? — lucasmeijer · 2026-10-22