Nous Research's NousCoder-14B is an open-source coding model landing right in the Claude Code moment

VentureBeat AI · 6 months ago

Nous Research released NousCoder-14B, a 14-billion-parameter open-source coding model trained in four days on 48 Nvidia B200 GPUs, achieving 67.87% accuracy on competitive programming benchmarks. The model was trained using reinforcement learning on 24,000 competitive programming problems with verifiable test case execution, representing a 7.08 percentage point improvement over its base model. The release arrives as Claude Code dominates social media, intensifying competition in AI-assisted coding, while Nous Research emphasizes reproducibility by publishing complete training infrastructure and confronts a looming data scarcity challenge in the competitive programming domain.