SeaEval - Multilingual Reasoning - IndoMMLU (Zero-Shot): leaderboard
Metric: Accuracy (%). Source: huggingface.co. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.1 70B Instruct | 67.41 |
| 2 | Qwen 2 72B Instruct | 63.86 |
| 3 | Qwen 2.5 72B Instruct | 63.81 |
| 4 | Llama 3 70B Instruct | 63.24 |
| 5 | Qwen 2.5 32B Instruct | 63.15 |
| 6 | gemma2-9B-cpt-sea-lionv3-instruct | 61.96 |
| 7 | Gemma 2 9B (IT) | 60.7 |
| 8 | Qwen 2.5 14B Instruct | 60.1 |
| 9 | Llama 3.1 8B Instruct | 56.05 |
| 10 | Qwen 2.5 7B Instruct | 56.01 |
| 11 | Qwen 2 7B Instruct | 53.86 |
| 12 | Llama 3 8B Cpt Sea Lionv2.1 Instruct | 52.69 |
| 13 | SeaLLMs-v3-7B-Chat | 52.67 |
| 14 | Llama 3 8B Instruct | 52.65 |
| 15 | Llama 3 8B Cpt Sea Lionv2 Instruct | 52.53 |
Interactive version: theaggregate.ai/benchmark?slug=seaeval-multilingual-reasoning-indommlu-zero-shot · How It Works · Data refreshed daily, snapshot 2026-09-05.