PDB-Wild - Unit Test Pass Rate: leaderboard
Metric: Unit-test pass rate (%) of the debugged programs on PDB-Wild multi-line and repository-level bugs; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 85.1 |
| 2 | GPT-5.5 | 84.3 |
| 3 | Claude Opus 4.7 | 75.8 |
| 4 | Claude Sonnet 4.5 (Thinking) | 69.2 |
| 5 | Gemini 2.5 Pro | 68.8 |
| 6 | GPT-5.1 Codex | 68.6 |
| 7 | Qwen 3.6 Plus | 65.3 |
| 8 | Kimi K2.6 | 43.4 |
Interactive version: theaggregate.ai/benchmark?slug=pdb-wild-unit-test-pass-rate · How It Works · Data refreshed daily, snapshot 2026-10-07.