Cursor's first-party harder-task benchmark for long-horizon agentic coding behavior inside the Cursor agent loop.
BenchLM mirrors the published score view for CursorBench v3.1. Composer 2.5 leads the public snapshot at 63.2%. BenchLM does not use these results to rank models overall.
Year
2026
Tasks
Harder long-horizon agentic coding tasks
Format
Cursor agent-loop evaluation
Difficulty
Professional agentic software engineering
Cursor reports CursorBench v3.1 alongside public Terminal-Bench 2.0 and SWE-bench Multilingual results in the Composer 2.5 launch table. BenchLM tracks it as display-only because it is a first-party benchmark.
Version
CursorBench v3.1 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Cursor's first-party harder-task benchmark for long-horizon agentic coding behavior inside the Cursor agent loop.
Composer 2.5 by Cursor currently leads with a score of 63.2% on CursorBench v3.1.
1 AI models have been evaluated on CursorBench v3.1 on BenchLM.
For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.