TEBench - Identification: leaderboard

Identification F1 (%) for all affected tests on TEBench (2026), a project-level test-evolution benchmark of 314 commits from 10 Java projects in which an agent in its default configuration updates the test suite after a production-code change. The tests the agent modifies, deletes and adds are matched against the test changes the developers made; higher is better.

Metric: Identification F1 (%) for all three evolution types: the tests the agent modified, deleted (method level) or added (file level) against the developer's own test changes, 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 March 2027. 6 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-5.3 Codex49.4#124
2GLM-549.3#269
3Kimi K2.548.5#266
4Claude Sonnet 4.648.3#173
5Qwen3.5 (TEBench checkpoint unspecified)48.2
6DeepSeek V3.245.7#419

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=tebench-identification · How It Works · Data refreshed daily, snapshot 2026-11-21.