CodeSpecBench-Repo: leaderboard
CodeSpecBench (2026) asks models to write executable precondition and postcondition functions that specify the intended behaviour of a program; the repository-level set uses the 500 SWE-bench Verified issues, injecting the specifications around the issue-relevant functions. Pass rate, the share of specifications that accept every valid test and reject every invalid one, in percent; higher is better.
Metric: Pass rate (%), the share of generated executable pre/postcondition specifications that are both correct (accept every valid test) and complete (reject every invalid test), on CodeSpecBench-Repo (500 SWE-bench Verified issues; specifications injected around the issue-relevant functions, trigger tests on the fixed and buggy versions with UTBoost-augmented tests); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 15 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Claude Sonnet 4.5 | Thinking | 20.2 | #261 |
| 2 | Gemini 2.5 Pro | 18.2 | #276 | |
| 3 | GPT-5 Mini | 9.6 | #364 | |
| 4 | GPT-5 | 8.6 | #178 | |
| 5 | DeepSeek V3.2 | Non-reasoning | 6.8 | #419 |
| 6 | QwQ-32B | 3.6 | #907 | |
| 7 | GPT-OSS-20B | 3.4 | #1183 | |
| 8 | GPT-OSS-120B | 3.2 | #704 | |
| 9 | Gemini 2.5 Flash | 2.8 | #463 | |
| 10 | Qwen 3 32B | 2 | #972 | |
| 11 | Qwen 3 14B | 1.2 | #1269 | |
| 12 | Qwen 3 8B | Non-reasoning | 0.8 | #1873 |
| 13 | Qwen 3 4B | 0.2 | #2348 | |
| 14 | Qwen 3 1.7B | 0 | #5860 | |
| 15 | Qwen 3 0.6B | 0 | #8350 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=codespecbench-repo · How It Works · Data refreshed daily, snapshot 2026-11-21.