TEBench - Update: leaderboard
Metric: Update overall composite (%) for all three evolution types: executability gate times a blend of coverage overlap with the developer tests on the changed methods and token-level modification similarity, on TEBench (314 commits from 10 Java projects): given the project at the commit with its production changes but not its test changes, the agent edits the test suite in an isolated worktree with default agent settings; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.3 Codex | 72.3 |
| 2 | GLM-5 | 69.3 |
| 3 | Claude Sonnet 4.6 | 68.9 |
| 4 | DeepSeek V3.2 | 64.5 |
| 5 | Kimi K2.5 | 63.6 |
Interactive version: theaggregate.ai/benchmark?slug=tebench-update · How It Works · Data refreshed daily, snapshot 2026-10-07.