Open Japanese LLM - AIO Char F1: leaderboard
The "AIO Char F1" 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 | Llama-3.3-Swallow-70B-Instruct-v0.4 | 89.25 | #529 |
| 2 | llm-jp-3-13B | 85.77 | #1219 |
| 3 | llm-jp-3.1-13B-instruct4 | 80.68 | #835 |
| 4 | aya-expanse-32B | 74.89 | #692 |
| 5 | ABEJA-Qwen2.5-32B-Japanese-v1.0 | 74.07 | #503 |
| 6 | Llama-3.1-Tulu-3-70B | 74.01 | #482 |
| 7 | llm-jp-3-3.7B | 74.01 | #1435 |
| 8 | Llama 3 70B | 73.47 | #631 |
| 9 | Llama 3.1 70B | 72.96 | #592 |
| 10 | Llama 3.1 Nemotron 70B Instruct | 72.7 | #554 |
| 11 | Llama 3.1 70B Instruct | 72.64 | #561 |
| 12 | aya-23-35B | 70.88 | #883 |
| 13 | Llama 3.3 70B Instruct | 70.84 | #534 |
| 14 | Gemma 2 27B (IT) | 69.27 | #645 |
| 15 | Llama 3 70B Instruct | 67.3 | #634 |
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-aio-char-f1 · How It Works · Data refreshed daily, snapshot 2026-10-20.