SOLAR-10.7B-v1.0: benchmark results
Upstage's Apache-2.0 10.7B pretrained base model (December 2023), built by depth up-scaling Mistral-7B weights on a Llama-2 architecture. Provider: Upstage. Released 2023-12-11. Access: Open.
Aggregate Rating 1391 ± 14, rank #1253 of 1656 rated models, from 369 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Romanian NLU - MultiWikiQA RO | 77.57 | Character F1 (%) | 99.5 |
| EuroEval Portuguese NLU - MultiWikiQA PT | 79.82 | Character F1 (%) | 99.3 |
| EuroEval Italian NLU - SQuAD IT | 77.48 | Character F1 (%) | 98.8 |
| EuroEval Catalan NLU - MultiWikiQA CA | 78.2 | Character F1 (%) | 98.6 |
| EuroEval Danish NLU - Multi Wiki QA DA | 82.34 | Character F1 (%) | 98.5 |
| EuroEval Spanish NLU - MLQA ES | 70.14 | Character F1 (%) | 98.5 |
| EuroEval Bulgarian NLU - MultiWikiQA BG | 79.17 | Character F1 (%) | 97.6 |
| EuroEval Slovak NLU - Multi Wiki QA SK | 73.95 | Character F1 (%) | 97.5 |
| EuroEval Polish NLU - PoQuAD | 63.61 | Character F1 (%) | 97 |
| EuroEval Swedish NLU - Multi Wiki QA SV | 80.55 | Character F1 (%) | 96.4 |
| EuroEval Slovene NLU - MultiWikiQA SL | 72.15 | Character F1 (%) | 96.1 |
| EuroEval Hungarian NLU - MultiWikiQA HU | 69.79 | Character F1 (%) | 96 |
Interactive version: theaggregate.ai/model?slug=solar-10-7b-v1-0 · How It Works · Data refreshed daily, snapshot 2026-10-20.