Chinese Language Understanding on CMMLU
90.9ScoreKimi-K2
Evaluation Results
| Method | Links | |
|---|---|---|
| Kimi-K2Model Variant=Base, # Shots=5-shot, # Activated Params=32B, # Total Params=1043B2026.02 | 90.9 | |
| Step 3.5 FlashModel Variant=Base, # Shots=5-shot, # Activated Params=11B, # Total Params=196B2026.02 | 88.9 | |
| DeepSeek V3.2Model Variant=Exp Base, # Shots=5-shot, # Activated Params=37B, # Total Params=671B2026.02 | 88.9 | |
| DeepSeek V3.1Model Variant=Base, # Shots=5-shot, # Activated Params=37B, # Total Params=671B2026.02 | 88.8 | |
| MiMo-V2 FlashModel Variant=Base, # Shots=5-shot, # Activated Params=15B, # Total Params=309B2026.02 | 87.4 | |
| NBDiff-7B-INSTRUCTParameters=7B, Training Protocol=Instruct, Sampling=Standard2025.12 | 79.8 | |
| SDAR 8BParameters=8B, Sampling=Standard2025.12 | 75.7 | |
| LLaDA2.0-mini preview 16BA1BParameters=16BA1B, Sampling=Standard2025.12 | 67.5 | |
| LLaDA-MoE 7B-A1BParameters=7B-A1B, Sampling=Standard2025.12 | 64.3 | |
| Dream-v0 Instruct-7BParameters=7B, Training Protocol=Instruct, Sampling=Standard2025.12 | 58.8 |