General Large Language Model Evaluation on Core Capabilities Aggregate
72.32Average ScoreQwen3-4B
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3-4BParams=4B2025.12 | 72.32 | |
| Qwen2.5-3BParams=3B2025.12 | 63.15 | |
| Qwen3-1.7BParams=1.7B2025.12 | 61.47 | |
| LLama-3-8BModel Size=8B, Tokens=15T2026.02 | 56.5 | |
| DCLM-Bas. + CC-TablesModel Size=7B, Tokens=2.7T, alpha=0.82026.02 | 56.1 | |
| DCLM-7B-8kModel Size=7B, Tokens=2.8T2026.02 | 56 | |
| Qwen2.5-1.5BParams=1.5B2025.12 | 55.95 | |
| DCLM-Bas. + CC-TablesModel Size=7B, Tokens=2.7T, alpha=0.852026.02 | 55.7 | |
| DCLM-Bas. + CC-TablesModel Size=7B, Tokens=2.7T, alpha=0.92026.02 | 54.7 | |
| DCLM-Bas. + CC-TablesModel Size=7B, Tokens=2.7T, alpha=12026.02 | 53.7 | |
| YuLan-Mini-2.4BParams=2.4B2025.12 | 53.02 | |
| SmolLM3-3BParams=3B2025.12 | 52.64 | |
| Qwen2-1.5BParams=1.5B2025.12 | 50.63 | |
| Llama-3.2-3BParams=3B2025.12 | 49.17 | |
| Qwen3-0.6BParams=0.6B2025.12 | 47.18 | |
| PCMind-2.1-Kaiyuan-2BParams=2B2025.12 | 46.05 | |
| Llama-3.2-1BParams=1B2025.12 | 31.56 | |
| SmolLM2-1.7BParams=1.7B2025.12 | 30.64 | |
| Gemma2-2BParams=2B2025.12 | 29.42 | |
| OLMo-2-0425-1BParams=1B2025.12 | 28.4 |