A 30B open model trained entirely on offline research trajectories beat GPT-4.1, Claude-4-Opus, and DeepSeek-R1 on BrowseComp-Plus. OpenResearcher scored 54.8%

A 30B open model trained entirely on offline research trajectories beat GPT-4.1, Claude-4-Opus, and DeepSeek-R1 on BrowseComp-Plus. OpenResearcher scored 54.8%, up 34 points from its base model. It also transferred to the live web without live-web training data. [📷](https://pbs.twimg.com/media/…