Cursor's proprietary coding model powering Composer agent mode. Tops CursorBench (61.3), Terminal-Bench (61.7), and SWE-bench Multilingual (73.7). Trained specifically for multi-file editing and agentic coding workflows.
| Signal | Strength | Weight | Impact |
|---|---|---|---|
| Capabilitiesjust now | 50 | 30% | +15.0 |
| Recencyjust now | 100 | 15% | +15.0 |
| Context Windowjust now | 84 | 15% | +12.6 |
| Output Capacityjust now | 80 | 15% | +12.0 |
| Pricingjust now | 3 | 25% | +0.6 |
Community and practitioner feedback adds real-world signal on top of benchmarks and pricing.
Share your experience with Composer 2 and help the community make better decisions.
Cost Estimator
You save $31.25/month vs category average