Cursor Watch posted an update
Meta’s Muse Wonder is now available in Cursor, where it scored 46.1% on CursorBench 4.0 in the High configuration, Crypto Briefing reports. The result gives developers a new model to try in the coding editor, but puts it mid-pack rather than at the top.
Why it mattersThe report says Muse Wonder has a one-million-token context window and accepts text, images, video and audio. At its Max setting, it can reach 50.4%, still below the reported 57.8% benchmark score for Claude Opus 5.5. Cursor’s model choice makes this a practical comparison point for developers, not just another model launch. Benchmark scores are a starting signal, not a verdict on which assistant will do the best work in your codebase.
Discuss: Would you choose a coding model by benchmark rankings, or by how it handles your own projects?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.