Language Understanding on MMLU v1 (test)
72.4AccuracyGLM-4-9B
Evaluation Results
| Method | Links | |
|---|---|---|
| GLM-4-9BInstruction-tuned=true2024.07 | 72.4 | |
| Qwen2-7BInstruction-tuned=true2024.07 | 70.5 | |
| Yi-1.5-9BInstruction-tuned=true2024.07 | 69.5 | |
| Llama-3-8BInstruction-tuned=true2024.07 | 68.4 | |
| LLaVA-70BMultimodal-Language Data Mix=true2023.09 | 65.1 | |
| LLaMA-2-70B-Chat2023.09 | 63.1 | |
| LLaVA-65BMultimodal-Language Data Mix=false2023.09 | 62.6 | |
| Vicuna-65B2023.09 | 62.5 | |
| LLaVA-65BMultimodal-Language Data Mix=true2023.09 | 62.2 | |
| Qwen1.5-7BInstruction-tuned=true2024.07 | 59.5 | |
| Vicuna-33B2023.09 | 59 | |
| LLaVA-33BMultimodal-Language Data Mix=true2023.09 | 58.6 | |
| LLaVA-33BMultimodal-Language Data Mix=false2023.09 | 56.1 | |
| Vicuna-13B2023.09 | 55.8 | |
| LLaVA-13BMultimodal-Language Data Mix=false2023.09 | 55 |