LongBench Pro - Easy - Non-Thinking: leaderboard
Metric: Score (%). Source: arxiv.org. 34 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.7 Sonnet | 68.69 |
| 2 | Claude Sonnet 4 | 68.42 |
| 3 | Gemini 2.5 Flash | 66.55 |
| 4 | DeepSeek V3 (0324) | 65.26 |
| 5 | Qwen 3 Next 80B A3B Instruct | 65.25 |
| 6 | Kimi K2 0905 | 64.92 |
| 7 | DeepSeek V3.1 | 63.61 |
| 8 | DeepSeek V3.2 | 62.12 |
| 9 | Qwen 3 235B A22B 2507 Instruct | 61.76 |
| 10 | GPT-4o | 59.38 |
| 11 | GLM-4.6 | 58.82 |
| 12 | Qwen 3 30B A3B 2507 Instruct | 55.89 |
| 13 | GLM-4.5 | 55.68 |
| 14 | Llama 3.1 405B Instruct | 55.45 |
| 15 | Qwen 2.5 72B Instruct | 53.48 |
Interactive version: theaggregate.ai/benchmark?slug=longbench-pro-easy-non-thinking · How It Works · Data refreshed daily, snapshot 2026-09-19.