Open Japanese LLM - Wikicorpus J TO E Bleu EN: leaderboard
The "Wikicorpus J TO E Bleu EN" 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: Around 2036. 418 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | llm-jp-3.1-13B-instruct4 | 15.86 | #835 |
| 2 | llm-jp-3.1-1.8B-instruct4 | 14.65 | #1273 |
| 3 | Llama-3.1-Tulu-3-70B | 13.1 | #482 |
| 4 | Llama 3.1 70B | 12.69 | #592 |
| 5 | Llama 3 70B | 12.58 | #631 |
| 6 | Llama 3.1 70B Instruct | 12.39 | #561 |
| 7 | Llama-3.3-Swallow-70B-Instruct-v0.4 | 12.37 | #529 |
| 8 | Llama 3.3 70B Instruct | 12.3 | #534 |
| 9 | Gemma 2 27B (IT) | 12.11 | #645 |
| 10 | Llama 3.1 Nemotron 70B Instruct | 12.08 | #554 |
| 11 | DeepSeek R1 Distill Llama 70B | 11.71 | #594 |
| 12 | Llama 2 70B Base | 11.71 | #1025 |
| 13 | Qwentile2.5-32B-Instruct | 11.63 | #339 |
| 14 | aya-expanse-32B | 11.56 | #692 |
| 15 | Qwen 2.5 32B | 11.5 | #546 |
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-j-to-e-bleu-en · How It Works · Data refreshed daily, snapshot 2026-10-20.