LongCat-Flash-Thinking-2509: benchmark results
Provider: Meituan. Access: API.
Aggregate Rating 1655 ± 22, rank #591 of 10916 rated models, from 29 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AGI-Eval Community - General Reasoning | 96.92 | Accuracy (%) | 70.7 |
| AGI-Eval Community - Instruction Following | 44.41 | Accuracy (%) | 66.4 |
| AGI-Eval Community - Learning (English) | 91.68 | Accuracy (%) | 62.3 |
| AGI-Eval Community - Algorithmic Reasoning | 39.67 | Accuracy (%) | 57.9 |
| AGI-Eval Community - Subject Reasoning (Chinese) | 83.75 | Accuracy (%) | 57.2 |
| AGI-Eval Community - Objective Accuracy (Chinese) | 62.31 | Accuracy (%) | 55.8 |
| AGI-Eval Community - Interaction (English) | 90.77 | Accuracy (%) | 55.1 |
| AGI-Eval Community - Objective Accuracy | 67.15 | Accuracy (%) | 55 |
| AGI-Eval Community - Subject Knowledge | 84.26 | Accuracy (%) | 54.6 |
| AGI-Eval Community - Subject Reasoning (English) | 82.68 | Accuracy (%) | 54.3 |
| AGI-Eval Community - Interaction | 80.29 | Accuracy (%) | 53.6 |
| AGI-Eval Community - Subject Reasoning | 83.01 | Accuracy (%) | 52.5 |
Interactive version: theaggregate.ai/model?slug=longcat-flash-thinking-2509 · How It Works · Data refreshed daily, snapshot 2026-11-08.