SciIntegrity-Bench: leaderboard
Metric: Scenarios with explicit misconduct (count of 33; three scenarios in each of 11 trap categories; minimal ReAct agent, temperature 0; human annotation against pre-specified violation criteria). Source: arxiv.org. Saturation forecast: Around June 2027. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 1 |
| 2 | GPT-5.2 | 2 |
| 3 | DeepSeek V3.2 | 3 |
| 4 | Gemini 3.1 Pro (Preview) | 5 |
| 5 | Qwen 3.5 397B A17B | 6 |
| 6 | GLM-5 | 7 |
| 7 | Kimi K2.5 | 12 |
Interactive version: theaggregate.ai/benchmark?slug=sciintegrity-bench · How It Works · Data refreshed daily, snapshot 2026-09-26.