Solar Mini 4: benchmark results
Provider: Upstage. Access: Open.
Unified ELO 1701 ± 40, rank #213 of 1609 rated models, from 58 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Long Context Reasoning | 83.33 | Accuracy (%) | 96.4 |
| AA-Omniscience Hallucination Rate | 35.83 | Hallucination Rate (%) | 89 |
| Artificial Analysis Intelligence Index | 24.05 | Intelligence Index | 75 |
| AA-Omniscience Index - Law | -7.5 | Omniscience Index | 74.2 |
| AA Humanity's Last Exam | 25.76 | Accuracy (%) | 74 |
| AA-Omniscience Index - Business | -8 | Omniscience Index | 73.3 |
| AA-Omniscience Index - Humanities & Social Sciences | -7.6 | Omniscience Index | 71.2 |
| AA-Omniscience Index - Science, Engineering & Mathematics | -6.1 | Omniscience Index | 70 |
| AA Omniscience | -10.82 | Score | 66.6 |
| AA-Omniscience Index - Health | -20.1 | Omniscience Index | 64.7 |
| AA CritPt | 1.43 | Accuracy (%) | 63.6 |
| AA-Omniscience Index - Software Engineering (SWE) - Dart | -16 | Omniscience Index | 63 |
Interactive version: theaggregate.ai/model?slug=solar-mini-4 · How It Works · Data refreshed daily, snapshot 2026-10-02.