Solar Pro 3: benchmark results
Upstage's enterprise reasoning MoE (102B total/12B active), Korean/English bilingual (January 2026). Provider: Upstage. Released 2026-01-27. Access: API.
Unified ELO 1547 ± 1, rank #453 of 1392 rated models, from 29 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA IFBench | 71.2 | Accuracy (%) | 86.9 |
| AA TAU-2 Bench | 86.26 | Accuracy (%) | 78.9 |
| AA Humanity's Last Exam | 10.29 | Accuracy (%) | 54.9 |
| UGI - Natural Intelligence | 23.04 | NatInt Score | 54.5 |
| AA GPQA Diamond | 72.42 | Accuracy (%) | 51.8 |
| AA Omniscience - Health | 20.1 | Accuracy (%) | 48.1 |
| AA Omniscience - Law | 11.2 | Accuracy (%) | 48.1 |
| Artificial Analysis Intelligence Index | 8.53 | Intelligence Index | 44.6 |
| BenchmarkList ECI | 105.52 | Capability Index (ECI) | 43.6 |
| AA-Omniscience Accuracy | 18.5 | Accuracy (%) | 42.5 |
| AA Omniscience - Software Engineering (SWE) | 22.5 | Accuracy (%) | 40.9 |
| AA Terminal-Bench Hard | 7.58 | Accuracy (%) | 40 |
Interactive version: theaggregate.ai/model?slug=solar-pro-3 · How It Works · Data refreshed daily, snapshot 2026-09-05.