Open Japanese LLM - Jsem Exact Match: leaderboard
The "Jsem Exact Match" subtask of Open Japanese LLM. A task from the Open Japanese LLM Leaderboard, evaluating models on Japanese-language knowledge, reasoning and understanding.
Metric: Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 418 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | llm-jp-3.1-13B-instruct4 | 96.21 | #835 |
| 2 | llm-jp-3.1-1.8B-instruct4 | 87.82 | #1273 |
| 3 | Qwentile2.5-32B-Instruct | 80.93 | #339 |
| 4 | QwQ 32B-Preview | 80.49 | #651 |
| 5 | DeepSeek-R1-Distill-Qwen-32B-Japanese | 80.24 | #638 |
| 6 | Llama-3.1-Tulu-3-70B | 80.18 | #482 |
| 7 | Llama 3.1 70B Instruct | 79.92 | #561 |
| 8 | Qwen2.5-32B-Instruct-CFT | 79.86 | #375 |
| 9 | Qwen 2.5 32B Instruct | 79.8 | #498 |
| 10 | QwQ-32B | 79.8 | #421 |
| 11 | Awqward2.5-32B-Instruct | 79.8 | #371 |
| 12 | Qwen2.5-32B-Instruct-abliterated-v2 | 79.73 | #394 |
| 13 | oxyge1-33B | 79.73 | #690 |
| 14 | Llama 3.1 70B | 79.61 | #592 |
| 15 | Rombos-LLM-V2.5-Qwen-32B | 79.61 | #367 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-japanese-llm-jsem-exact-match · How It Works · Data refreshed daily, snapshot 2026-10-20.