Models & Labs
Recursive AI Launches Flint RSI Coding Model
Sam Witteveenhigh confidence
Why it matters
- →Flint offers open weights with NVFP4 quantization, enabling high-speed coding inference on consumer GPUs.
- →It introduces Recursive AI's RSI framework to the public, providing a new architectural approach for MoE models.
- →The model directly benchmarks against Qwen3.8 27B, establishing a new baseline for local coding assistants.
Recursive AI has released Flint, its first public mixture-of-experts model designed specifically for coding tasks and built using its Recursive System Intelligence (RSI) framework. The release includes four open-weight versions, including an NVFP4 quantized variant optimized for efficient inference on hardware like the NVIDIA RTX PRO 6000. Benchmarks position Flint as a direct competitor to Qwen3.8 27B, emphasizing speed and local deployability through support for Ollama, LM Studio, and vLLM. The model was developed by Recursive AI's 'Eureka Machine' in under four months using automated architecture search.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
More from Sam Witteveen
More in Models & Labs
Models & Labsmodels
llama.cpp b12029 adds CUDA 13 and Snapdragon support
This release quietly cements llama.cpp as the universal inference runtime by finally supporting NVIDIA's latest CUDA 13 stack alongside AMD's ROCm 10.0. The inclusion of native builds for Linux arm64 with Snapdragon chips, covering CPU, Adreno GPU, and Hexagon NPU, signals a serious push into mobile edge deployment. While Apple Silicon KleidiAI is disabled in this specific build, the broader hardware coverage means developers no longer need to compile from source for most modern accelerators. It’s less about new features and more about removing friction for anyone running local models on bleeding-edge or diverse hardware.
llama.cpp Releases
© The Verge AIModels & Labsother
