Long Context Understanding on MRCR
75.3AccuracyGemini-3.0-pro
Evaluation results
| Method | Links | |
|---|---|---|
| Gemini-3.0-pro2026.03 | 75.3 | |
| LongCat-Flash Exp-ChatEvaluation Mode=Chat2025.12 | 59.7 | |
| Deepseek-v3.12026.03 | 46.62 | |
| Qwen3-32B + TableLong2026.03 | 42.66 | |
| Qwen3-32B2026.03 | 42.45 | |
| GLM 4.6Evaluation Mode=Chat2025.12 | 42.1 | |
| Deepseek-R1-Distill-Qwen-32B + TableLong2026.03 | 40.57 | |
| DeepSeek V3.2Evaluation Mode=Chat2025.12 | 37.1 | |
| LongCat-Flash ChatEvaluation Mode=Chat2025.12 | 34.4 | |
| Qwen2.5-32B-Instruct2026.03 | 33.19 | |
| Qwen2.5-32B-Instruct + TableLong2026.03 | 33.19 | |
| Deepseek-R1-Distill-Qwen-32B2026.03 | 31.94 | |
| Deepseek-R1-Distill-Qwen-14B + TableLong2026.03 | 30.48 | |
| Deepseek-R1-Distill-Qwen-14B2026.03 | 29.02 | |
| Qwen-Long-L12026.03 | 27.7 |