CRAB-Bench: leaderboard
Metric: Pass rate (%): share of tasks passing both concrete-state (database) and abstract-state (conversation factuality and completion) verification, with the RUSE human-aligned user simulator (personas sampled uniformly, all four behavior dimensions); 200 trip-booking tasks (departure and return flight, hotel, attraction) generated from constraint graphs with node and edge distractors, tau-bench framework, pass@1, default inference and reasoning settings, Claude Sonnet 4.6 simulates the user; higher is better. Source: arxiv.org. Saturation forecast: Around May 2027. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V3.2 | 61 |
| 2 | GLM-5 | 44 |
| 3 | Claude Sonnet 4.6 | 32 |
| 4 | Qwen3 Coder Next | 20 |
Interactive version: theaggregate.ai/benchmark?slug=crab-bench · How It Works · Data refreshed daily, snapshot 2026-09-29.