XRepoTest - Go - Test Pass Rate: leaderboard
XRepoTest (arXiv 2608.25939v1, Appendix Table A.4): repository-level unit-test generation in real open-source Go projects, standard single-turn setting with greedy decoding, with generated tests executed in containers. Share of all focal functions whose generated test suite executes and passes, percent. Higher is better.
Metric: Passing generated test suites (%; of all focal functions). Source: arxiv.org. Saturation forecast: Around July 2028. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 25.78 | #115 |
| 2 | Claude Sonnet 4.5 | 23.65 | #148 |
| 3 | GPT-OSS-120B | 19.26 | #346 |
| 4 | DeepSeek V4 Pro | 13.88 | #112 |
| 5 | Claude Haiku 4.5 (20251001) | 10.91 | #264 |
| 6 | Qwen3 Coder Next | 6.66 | #331 |
| 7 | DeepSeek V3.2 | 5.24 | #210 |
| 8 | GLM-5 | 5.1 | #149 |
| 9 | MiniMax-M2.7 | 4.67 | #270 |
| 10 | Codestral-22B-v0.1 | 3.82 | #850 |
| 11 | Llama 3.3 70B Instruct | 2.69 | #541 |
| 12 | Yi-Coder-9B-Chat | 2.41 | #1244 |
| 13 | Kimi K2.5 | 2.27 | #152 |
| 14 | Qwen 3 8B | 0.42 | #678 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=xrepotest-go-test-pass-rate · How It Works · Data refreshed daily, snapshot 2026-11-05.