Flint Local Test | The Qwen3.8 35B MoE We Need? | Coding, Frontend, Backend | Local AI Setup

Recursive Flint is 40B MoE (2.5B active) parameter model that should (in benchmaxing terms) satisfy our need for a new Qwen MoE model. I'll run it locally, evaluate it on DaBench and see if this is a worthy successor? DaBench run: https://dabench.ai/model/flint-v-1-0 Announcement: https://recursive.com/updates/introducing-recursive-flint Weights: https://huggingface.co/RecursiveSI/flint-nvfp4-gguf AI Academy: https://MLExpert.io Work with me: https://mlexpert.io/consulting DaBench (open weight LLM benchmarks): https://dabench.ai LinkedIn: https://www.linkedin.com/in/venelin-valkov Follow me on