Open Japanese LLM - Wikicorpus E TO J Bleu JA: leaderboard
The "Wikicorpus E TO J Bleu JA" 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 | 24.01 | #835 |
| 2 | llm-jp-3.1-1.8B-instruct4 | 20.47 | #1273 |
| 3 | Llama 3.1 Nemotron 70B Instruct | 16.75 | #554 |
| 4 | Llama 3.1 70B Instruct | 16.63 | #561 |
| 5 | Llama 3 70B | 16.12 | #631 |
| 6 | Llama 3.3 70B Instruct | 15.86 | #534 |
| 7 | Llama 3.1 70B | 15.48 | #592 |
| 8 | Llama-3.1-Tulu-3-70B | 15.38 | #482 |
| 9 | Llama 2 70B Base | 14.17 | #1025 |
| 10 | Qwen 2.5 32B | 13.7 | #546 |
| 11 | Llama-3.3-Swallow-70B-Instruct-v0.4 | 13.52 | #529 |
| 12 | aya-expanse-32B | 13.02 | #692 |
| 13 | DeepSeek R1 Distill Llama 70B | 13.02 | #594 |
| 14 | Qwentile2.5-32B-Instruct | 12.95 | #339 |
| 15 | Gemma 2 27B (IT) | 12.45 | #645 |
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-wikicorpus-e-to-j-bleu-ja · How It Works · Data refreshed daily, snapshot 2026-10-20.