IQuest Coder

Description

IQuest Coder is an open-source, code-specialized large language model that uses AI innovations—Code-Flow training, reinforcement learning for reasoning, and a Loop recurrent architecture—to understand repository evolution, perform iterative self-review, and generate, test, and refine multi-file code with native 128K-context support; available in 7B/14B/40B sizes and optimized Instruct and Thinking variants, it accelerates end-to-end software engineering (from design to debugging) with state-of-the-art benchmark performance, lower compute needs for local or cloud deployment, and extensibility for teams that want powerful, customizable AI coding assistance.