Nejumi 4 - jaster (2-shot) - KUCI: leaderboard
KUCI, Kyoto University commonsense inference with multiple-choice continuations, 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: Estimated already saturated. 145 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 94 | #64 |
| 2 | GPT-5.6 Sol (Max) | 93 | #22 (GPT-5.6 Sol) |
| 3 | Claude Opus 4.6 (Thinking) | 93 | #67 (Claude Opus 4.6) |
| 4 | Gemini 3 Flash (Preview) | 92 | #90 |
| 5 | Claude Opus 4.7 (xHigh) | 92 | #51 (Claude Opus 4.7) |
| 6 | Gemini 3 Pro (Preview) | 91 | #71 |
| 7 | Gemini 3.6 Flash | 91 | #57 |
| 8 | Claude Opus 5 (Max) | 91 | #16 (Claude Opus 5) |
| 9 | Claude Opus 4.5 (Thinking) | 91 | #87 (Claude Opus 4.5) |
| 10 | Muse Spark Open (xHigh) | 91 | #42 (Muse Spark Open) |
| 11 | Gemini 3.5 Flash | 90 | #66 |
| 12 | GPT-5.4 (2026-03-05) (High) | 90 | #63 (GPT-5.4 (2026-03-05)) |
| 13 | DeepSeek V4.1 Flash | 89 | #37 |
| 14 | Kimi K3 (Max) | 89 | #35 (Kimi K3) |
| 15 | GLM-5.3 Flash (Max) | 89 | #77 (GLM-5.3 Flash) |
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-kuci · How It Works · Data refreshed daily, snapshot 2026-11-05.