General Tasks on MMLU-Pro
82.93AccuracyQwen3-Next-80B-A3B-Instruct
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-Next-80B-A3B-InstructArchitecture=MoE, # Total Params=80B, # Activated Params=3B2026.01 | 82.93 | |
| Gemini 2.5 Flash-Lite2026.01 | 78.95 | |
| LongCat-Flash-LiteArchitecture=MoE + NE, # Total Params=68.5B, # Activated Params=2.9B~4.5B2026.01 | 78.29 | |
| Llama-3.1-405BModel Type=Instruct, Parameters=405B2025.02 | 73.3 | |
| Qwen2.5-VL-72BModel Type=Instruct, Parameters=72B2025.02 | 71.2 | |
| Qwen2.5-72BModel Type=Instruct, Parameters=72B2025.02 | 71.1 | |
| Kimi-Linear-48B-A3BArchitecture=MoE, # Total Params=48B, # Activated Params=3B2026.01 | 67.22 | |
| Llama-3.1-70BModel Type=Instruct, Parameters=70B2025.02 | 66.4 | |
| PODSModel=Qwen3-8B, Selection Ratio=50%2026.05 | 66.4 | |
| Full DataModel=Qwen3-8B, Selection Ratio=100%2026.05 | 65.8 | |
| Qwen2-72BModel Type=Instruct, Parameters=72B2025.02 | 64.4 | |
| PODSModel=Qwen3-4B, Selection Ratio=50%2026.05 | 61.5 | |
| Full DataModel=Qwen3-4B, Selection Ratio=100%2026.05 | 61.1 | |
| Fine-tunedBase LLM=Qwen-2.5-7B-Instruct, Parameters=x42026.05 | 55.6 | |
| DiDi-Merg.-LBase LLM=Qwen-2.5-7B-Instruct, Parameters=x2.02026.05 | 55.2 | |
| FREE-MergingBase LLM=Qwen-2.5-7B-Instruct, Parameters=x2.082026.05 | 54.8 | |
| Twin-MergingBase LLM=Qwen-2.5-7B-Instruct, Parameters=x2.252026.05 | 54.5 | |
| Zero-shotBase LLM=Qwen-2.5-7B-Instruct, Parameters=x12026.05 | 54 | |
| Fine-tunedBase LLM=Llama-3.1-8B-Instruct, Parameters=x42026.05 | 51.3 | |
| DiDi-Merg.-LBase LLM=Llama-3.1-8B-Instruct, Parameters=x2.02026.05 | 51.1 | |
| FREE-MergingBase LLM=Llama-3.1-8B-Instruct, Parameters=x2.082026.05 | 50.4 | |
| Twin-MergingBase LLM=Llama-3.1-8B-Instruct, Parameters=x2.252026.05 | 50.2 | |
| Zero-shotBase LLM=Llama-3.1-8B-Instruct, Parameters=x12026.05 | 49.7 |