Nejumi 4 - jaster (2-shot) - JCoLA (out-of-domain): leaderboard
JCoLA out-of-domain split, Japanese grammatical acceptability judgments, from the llm-jp-eval jaster suite as run 2-shot by Nejumi Leaderboard 4 (Weights & Biases Japan). Score is exact match (%) on Nejumi's sampled test split; higher is better.
Metric: Exact match (%). Source: nejumi.ai. Saturation forecast: Around February 2027. 145 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V4 Flash (0731) (Max) | 91 | #84 (DeepSeek V4 Flash (0731)) |
| 2 | Gemma 4 31B (IT) | 90 | #178 |
| 3 | GPT-5.6 Terra (Max) | 90 | #40 (GPT-5.6 Terra) |
| 4 | Claude Opus 4.6 (Thinking) | 89 | #67 (Claude Opus 4.6) |
| 5 | GPT-5.4 (2026-03-05) (xHigh) | 89 | #63 (GPT-5.4 (2026-03-05)) |
| 6 | GPT-5.4 (2026-03-05) (High) | 89 | #63 (GPT-5.4 (2026-03-05)) |
| 7 | O1 (2024-12-17) (High) | 89 | #157 (O1 (2024-12-17)) |
| 8 | DeepSeek-V4-Flash-Vision-Exp (Max) | 89 | #88 (DeepSeek-V4-Flash-Vision-Exp) |
| 9 | Gemini 3 Flash (Preview) | 88 | #90 |
| 10 | GPT-5.6 Sol (Max) | 88 | #22 (GPT-5.6 Sol) |
| 11 | GPT-5.6 Luna (Max) | 88 | #79 (GPT-5.6 Luna) |
| 12 | GPT-5.2 (2025-12-11) (xHigh) | 88 | |
| 13 | DeepSeek V4.1 Flash | 87 | #37 |
| 14 | Claude Opus 4.5 (Thinking) | 87 | #87 (Claude Opus 4.5) |
| 15 | Qwen 3.6 Max (Preview) (Thinking) | 87 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=nejumi-4-jaster-2-shot-jcola-out-of-domain · How It Works · Data refreshed daily, snapshot 2026-11-05.