Open Japanese LLM - Jsick Exact Match: leaderboard
The "Jsick 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 | 92.55 | #835 |
| 2 | llm-jp-3.1-1.8B-instruct4 | 90.05 | #1273 |
| 3 | Llama 3.1 Nemotron 70B Instruct | 86.62 | #554 |
| 4 | Qwen2.5-32B-Instruct-abliterated-v2 | 86.54 | #394 |
| 5 | Qwen2.5-32B-Instruct-CFT | 85.87 | #375 |
| 6 | Rombos-LLM-V2.5-Qwen-32B | 85.79 | #367 |
| 7 | oxyge1-33B | 85.73 | #690 |
| 8 | Qwen 2.5 32B Instruct | 85.69 | #498 |
| 9 | EZO-Qwen2.5-32B-Instruct | 85.51 | #471 |
| 10 | Awqward2.5-32B-Instruct | 85.24 | #371 |
| 11 | Qwen2.5-14B-Instruct-1M | 84.84 | #578 |
| 12 | qwen2.5-bakeneko-32B-instruct-v2 | 84.78 | #562 |
| 13 | Qwen2.5-7B-Instruct-1M | 84.29 | #894 |
| 14 | Llama 3.1 70B Instruct | 83.82 | #561 |
| 15 | DeepSeek R1 Distill Llama 70B | 83.74 | #594 |
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-jsick-exact-match · How It Works · Data refreshed daily, snapshot 2026-10-20.