Benchmark
ZebraLogic
ZebraLogic is an evaluation framework for assessing large language models' logical reasoning capabilities through logic grid puzzles derived from…
- Modality
- text
- Categories
- reasoning
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 8
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Kimi K2 Instruct | Moonshot AI | 0.89 | 2025-07-11 | evidence |
| Kimi K2-Instruct-0905 | Moonshot AI | 0.89 | 2025-09-05 | evidence |
| LongCat-Flash-Chat | Meituan | 0.893 | 2025-08-29 | evidence |
| LongCat-Flash-Thinking | Meituan | 0.955 | 2025-09-22 | evidence |
| MiniMax M1 40K | MiniMax | 0.801 | 2025-06-16 | evidence |
| MiniMax M1 80K | MiniMax | 0.868 | 2025-06-16 | evidence |
| Qwen3 VL 235B A22B Thinking | Qwen | 0.973 | 2025-09-22 | evidence |
| Qwen3-235B-A22B-Instruct-2507 | Qwen | 0.95 | 2025-07-22 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.