# How is CursorBench scored?

Cursor hasn't published the correctness rubric; it reports accuracy averaged over repeated attempts at each task, alongside efficiency numbers most benchmarks leave out: completion tokens, end-to-end latency, and inference cost per task, tracked separately because a model that's barely more accurate but far slower or more expensive isn't more useful in an editor where a developer is waiting on the response. Models are also compared across several reasoning-effort settings, high, medium, and low, rather than at one fixed setting, since the same model can land at different points on Cursor's accuracy-versus-cost curve depending on how hard it's told to think.

That's a different scoring philosophy from [scoring an agent on whether every repeated attempt succeeds](/answers/reliabilitybench/how-do-you-measure-whether-an-agent-succeeds-consistently), which counts a task only if the model never fails it. Averaging rewards a model that's usually right over one that's inconsistent but occasionally perfect, which fits Cursor's own framing of the benchmark as a signal for interactive developer workflows rather than a strict reliability test.

---

Sources:
- Composer 2 Technical Report (arXiv:2603.24477): https://arxiv.org/html/2603.24477v2 (fetched 2026-09-15)

Source: https://tessary.ai/answers/cursorbench/how-is-cursorbench-scored
More on Cursorbench: https://tessary.ai/answers/cursorbench
From Tessary, agent reliability for AI agents in production: https://tessary.ai