Open Japanese LLM - Commonsensemoralja Exact Match: leaderboard
The "Commonsensemoralja Exact Match" 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 | 93.84 | #835 |
| 2 | ABEJA-Qwen2.5-32B-Japanese-v1.0 | 92.69 | #503 |
| 3 | Llama-3.3-Swallow-70B-Instruct-v0.4 | 91.68 | #529 |
| 4 | Llama-3.1-Tulu-3-70B | 91.41 | #482 |
| 5 | QwQ 32B-Preview | 91.26 | #651 |
| 6 | EZO-Qwen2.5-32B-Instruct | 91.23 | #471 |
| 7 | Llama 3.3 70B Instruct | 91.21 | #534 |
| 8 | cogito-v1-preview-qwen-14B | 90.96 | |
| 9 | qwen2.5-bakeneko-32B-instruct-v2 | 90.78 | #562 |
| 10 | Llama 3.1 70B | 90.63 | #592 |
| 11 | Qwentile2.5-32B-Instruct | 90.58 | #339 |
| 12 | Llama 3.1 Nemotron 70B Instruct | 90.51 | #554 |
| 13 | Qwen2.5-32B-Instruct-abliterated-v2 | 90.51 | #394 |
| 14 | Lamarckvergence-14B | 90.48 | #545 |
| 15 | Qwen 2.5 32B Instruct | 90.46 | #498 |
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-commonsensemoralja-exact-match · How It Works · Data refreshed daily, snapshot 2026-10-20.