Drift-Bench (State-Oriented, No Clarification) - Intention: leaderboard
Drift-Bench (Bao et al., University of Notre Dame, 2026) state-oriented set: 200 AgentBench operating-system and database tasks that at least two of three reference models solve with the clean instruction, rewritten so the instruction carries a changed intention (the stated goal differs from the one the user holds). The agent must act without being able to ask the user. Success rate in percent, mean of three trials; higher is better.
Metric: Success rate (%; 200 AgentBench operating-system and database tasks with an intention fault (the stated goal changed), no clarification tools, mean of 3 trials). Source: arxiv.org. Saturation forecast: Estimated already saturated. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 235B A22B 2507 Instruct | 56.16 | #619 |
| 2 | GLM-4.7 | 50.67 | #365 |
| 3 | GPT-5.2 | 50 | #172 |
| 4 | DeepSeek V3.2 | 49 | #403 |
| 5 | Gemini 2.5 Flash | 48.17 | #452 |
| 6 | GPT-OSS-120B | 47.17 | #699 |
| 7 | Llama 4 Maverick | 32.33 | #1105 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=drift-bench-state-oriented-no-clarification-intention · How It Works · Data refreshed daily, snapshot 2026-11-12.