Multilingual Knowledge on MMMLU
87.2AccuracyGLM 4.6
Evaluation Results
| Method | Links | |
|---|---|---|
| GLM 4.6Evaluation Mode=Chat2025.12 | 87.2 | |
| DeepSeek V3.2Evaluation Mode=Chat2025.12 | 86.7 | |
| LongCat-Flash Exp-ChatEvaluation Mode=Chat2025.12 | 85.2 | |
| LongCat-Flash ChatEvaluation Mode=Chat2025.12 | 81.7 | |
| Qwen3-4BTotal Parameters=4B, Active Parameters=4B, Trained Tokens=36T2025.11 | 60.67 | |
| Granite-4.0-HTotal Parameters=7B, Active Parameters=1B, Trained Tokens=15T2025.11 | 56.13 | |
| LFM2-2.6BTotal Parameters=2.6B, Active Parameters=2.6B, Trained Tokens=11T2025.11 | 55.39 | |
| LFM2-8B-A1BTotal Parameters=8.3B, Active Parameters=1.5B, Trained Tokens=13T2025.11 | 55.26 | |
| Gemma-3-4BTotal Parameters=4B, Active Parameters=4B, Trained Tokens=4T2025.11 | 50.14 | |
| SmolLM3-3BTotal Parameters=3.1B, Active Parameters=3.1B, Trained Tokens=11T2025.11 | 50.02 | |
| Llama-3.2-3BTotal Parameters=3.2B, Active Parameters=3.2B, Trained Tokens=9T2025.11 | 47.92 | |
| LFM2-1.2B# Total Params=1.2B, # Trained Tokens=11T2025.11 | 46.73 | |
| Qwen3-1.7B# Total Params=1.7B, # Trained Tokens=36T2025.11 | 46.51 | |
| LFM2-700M# Total Params=0.70B, # Trained Tokens=11T2025.11 | 43.28 | |
| Llama-3.2-1B# Total Params=1.2B, # Trained Tokens=9T2025.11 | 38.15 | |
| LFM2-350M# Total Params=0.35B, # Trained Tokens=11T2025.11 | 37.99 | |
| Gemma-3-1B# Total Params=1B, # Trained Tokens=2T2025.11 | 34.43 | |
| Qwen3-0.6B# Total Params=0.6B, # Trained Tokens=36T2025.11 | 30.84 |