Nejumi 4 - jaster (2-shot) - JCommonsenseQA: leaderboard
Metric: Exact match (%). Source: nejumi.ai. 137 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 100 |
| 2 | Gemini 3 Pro (Preview) | 100 |
| 3 | GPT-5 (High) | 100 |
| 4 | GPT-5.5 (xHigh) | 100 |
| 5 | GPT-5.6 Sol (Max) | 100 |
| 6 | GPT-5.1 (High) | 100 |
| 7 | GPT-5.6 Terra (Max) | 100 |
| 8 | Claude Opus 4.5 (Thinking) | 100 |
| 9 | GLM-5.3 (Max) | 100 |
| 10 | Claude Opus 4.6 (Thinking) | 100 |
| 11 | Llama-3.3-Swallow-70B-Instruct-v0.4 | 100 |
| 12 | GLM-5.3 Flash (Max) | 100 |
| 13 | MiMo V2.5 Pro (Thinking) | 100 |
| 14 | GPT-4.1 Mini | 99 |
| 15 | Gemini 2.5 Pro | 99 |
Interactive version: theaggregate.ai/benchmark?slug=nejumi-4-jaster-2-shot-jcommonsenseqa · How It Works · Data refreshed daily, snapshot 2026-09-19.