LibEvoBench: leaderboard
Large software projects often depend on older versions of libraries, even as APIs continue to evolve across releases. This creates a challenge for LLMs: they must maintain knowledge of multiple API versions, not merely the latest or most common one. However, current LLMs are trained on temporally mixed corpora and lack explicit mechanisms for such version-specific reasoning, leading to anachronist
Metric: SEUS (self-reported). Source: benchmarklist.com. Saturation forecast: Around July 2027. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 86 | #79 |
| 2 | GPT-5.5 | 85.1 | #31 |
| 3 | Claude Sonnet 4.6 | 81 | #90 |
| 4 | Claude Sonnet 4 | 77.6 | #199 |
| 5 | Gemini 3 Flash | 77.3 | #99 |
| 6 | GPT-5.1 | 70.7 | #135 |
| 7 | GPT-4.1 | 70.6 | #248 |
| 8 | GPT-5 | 70 | #98 |
| 9 | Gemini 2.5 Flash | 68.9 | #245 |
| 10 | Qwen 3.5 397B A17B | 67.9 | #146 |
| 11 | Gemini 2.0 Flash | 66 | #342 |
| 12 | Qwen 3.5 122B A10B | 60.3 | #173 |
| 13 | Qwen 3.5 35B A3B | 52.6 | #256 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=libevobench · How It Works · Data refreshed daily, snapshot 2026-10-20.