Drift-Bench (State-Oriented, With Clarification) - Intention: leaderboard
Metric: Success rate (%; 200 AgentBench operating-system and database tasks with an intention fault (the stated goal changed), clarification tools answered by a persona user simulator, mean of 3 trials). Source: arxiv.org. Saturation forecast: Around 2029. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 235B A22B 2507 Instruct | 66.67 |
| 2 | Llama 4 Maverick | 59.33 |
| 3 | GPT-5.2 | 55.9 |
| 4 | Gemini 2.5 Flash | 55.33 |
| 5 | DeepSeek V3.2 | 55.33 |
| 6 | GLM-4.7 | 52.17 |
| 7 | GPT-OSS-120B | 50.49 |
Interactive version: theaggregate.ai/benchmark?slug=drift-bench-state-oriented-with-clarification-intention · How It Works · Data refreshed daily, snapshot 2026-09-25.