CoderCup - Correctness: leaderboard
CoderCup slice measuring coding-agent correctness on the current multi-phase contest task.
Metric: Correctness (%). Source: codercup.ai. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Kimi | 89.2 | |
| 2 | Claude Code | 85.27 | |
| 3 | Codex | 85.27 | |
| 4 | Anti-Gravity | 83.82 |
Interactive version: theaggregate.ai/benchmark?slug=codercup-correctness · How It Works · Data refreshed daily, snapshot 2026-10-20.