Open Japanese LLM - Wikicorpus J TO E Comet Wmt22: leaderboard
The "Wikicorpus J TO E Comet Wmt22" 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 | 79.02 | #835 |
| 2 | Llama-3.1-Tulu-3-70B | 77.96 | #482 |
| 3 | llm-jp-3.1-1.8B-instruct4 | 77.56 | #1273 |
| 4 | Llama-3.3-Swallow-70B-Instruct-v0.4 | 77.15 | #529 |
| 5 | Llama 3 70B Instruct | 76.85 | #634 |
| 6 | Gemma 2 27B (IT) | 76.81 | #645 |
| 7 | Llama 3.1 70B | 76.7 | #592 |
| 8 | qwen2.5-bakeneko-32B-instruct-v2 | 76.47 | #562 |
| 9 | DeepSeek R1 Distill Llama 70B | 76.42 | #594 |
| 10 | Qwentile2.5-32B-Instruct | 76.31 | #339 |
| 11 | aya-expanse-32B | 76.21 | #692 |
| 12 | EZO-Qwen2.5-32B-Instruct | 76.16 | #471 |
| 13 | QwQ 32B-Preview | 76.12 | #651 |
| 14 | tempmotacilla-cinerea-0308 | 76.1 | #395 |
| 15 | Awqward2.5-32B-Instruct | 75.99 | #371 |
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-comet-wmt22 · How It Works · Data refreshed daily, snapshot 2026-10-20.