Open Japanese LLM - Wiki Coreference SET F1: leaderboard
The "Wiki Coreference 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 | 40.14 | #835 |
| 2 | llm-jp-3.1-1.8B-instruct4 | 28.33 | #1273 |
| 3 | Llama-3.1-Tulu-3-70B | 9.99 | #482 |
| 4 | Llama 3.1 Nemotron 70B Instruct | 9.84 | #554 |
| 5 | Qwen 2.5 14B Instruct | 9.69 | #647 |
| 6 | ABEJA-Qwen2.5-32B-Japanese-v1.0 | 8.53 | #503 |
| 7 | Saka-14B | 8.22 | #344 |
| 8 | Llama 3.1 70B Instruct | 8.08 | #561 |
| 9 | tempmotacilla-cinerea-0308 | 7.04 | #395 |
| 10 | lambda-qwen2.5-14B-dpo-test | 6.85 | #481 |
| 11 | Lamarckvergence-14B | 6.69 | #545 |
| 12 | Llama-3.3-Swallow-70B-Instruct-v0.4 | 6.14 | #529 |
| 13 | Awqward2.5-32B-Instruct | 6.09 | #371 |
| 14 | Qwen 2.5 Coder 14B Instruct | 5.92 | #675 |
| 15 | Llama 3.1 70B | 5.65 | #592 |
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-coreference-set-f1 · How It Works · Data refreshed daily, snapshot 2026-10-20.