Open Japanese LLM - Jamp Exact Match: leaderboard
The "Jamp 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 | 99.71 | #835 |
| 2 | llm-jp-3.1-1.8B-instruct4 | 93.68 | #1273 |
| 3 | Qwen2.5-32B-Instruct-abliterated-v2 | 71.55 | #394 |
| 4 | qwen2.5-bakeneko-32B-instruct-v2 | 70.69 | #562 |
| 5 | Llama-3.1-Tulu-3-70B | 69.54 | #482 |
| 6 | EZO-Qwen2.5-32B-Instruct | 69.25 | #471 |
| 7 | ABEJA-Qwen2.5-32B-Japanese-v1.0 | 69.25 | #503 |
| 8 | Qwentile2.5-32B-Instruct | 68.97 | #339 |
| 9 | Llama 3 70B Instruct | 68.39 | #634 |
| 10 | Llama-3.3-Swallow-70B-Instruct-v0.4 | 68.1 | #529 |
| 11 | Rombos-LLM-V2.5-Qwen-32B | 67.53 | #367 |
| 12 | Llama 3.3 70B Instruct | 67.24 | #534 |
| 13 | Qwen 2.5 32B Instruct | 67.24 | #498 |
| 14 | oxyge1-33B | 67.24 | #690 |
| 15 | Awqward2.5-32B-Instruct | 67.24 | #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-jamp-exact-match · How It Works · Data refreshed daily, snapshot 2026-10-20.