Open Japanese LLM - Wiki Dependency SET F1: leaderboard
The "Wiki Dependency SET 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 | llm-jp-3.1-13B-instruct4 | 88.2 | #835 |
| 2 | llm-jp-3.1-1.8B-instruct4 | 85.65 | #1273 |
| 3 | Llama-3.3-Swallow-70B-Instruct-v0.4 | 48.15 | #529 |
| 4 | Llama-3.1-Tulu-3-70B | 46.35 | #482 |
| 5 | Llama 2 70B Chat (HF) | 45.02 | #1264 |
| 6 | Linkbricks-Horizon-AI-Korean-Superb-22B | 43.14 | #708 |
| 7 | qwen2.5-bakeneko-32B-instruct-v2 | 42.16 | #562 |
| 8 | Qwen2.5-14B-Instruct-1M | 41.23 | #578 |
| 9 | aya-expanse-32B | 40.33 | #692 |
| 10 | Qwentile2.5-32B-Instruct | 39.16 | #339 |
| 11 | magnum-v2.5-12B-kto | 39.07 | #873 |
| 12 | Awqward2.5-32B-Instruct | 38.82 | #371 |
| 13 | Rombos-LLM-V2.5-Qwen-32B | 38.45 | #367 |
| 14 | cogito-v1-preview-qwen-14B | 38.42 | |
| 15 | QwQ 32B-Preview | 38.42 | #651 |
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-wiki-dependency-set-f1 · How It Works · Data refreshed daily, snapshot 2026-10-20.