CRAB-Bench - Concrete-State Verification: leaderboard
Metric: Concrete-state verification pass rate (%): share of tasks whose final booking and wallet databases satisfy every task-specific verifier, 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 December 2026. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V3.2 | 73 |
| 2 | Claude Sonnet 4.6 | 62 |
| 3 | GLM-5 | 56 |
| 4 | Qwen3 Coder Next | 28 |
Interactive version: theaggregate.ai/benchmark?slug=crab-bench-concrete-state-verification · How It Works · Data refreshed daily, snapshot 2026-09-29.