Sonar: benchmark results
Perplexity's search-grounded API model built on Llama 3.3 70B and served on Cerebras hardware for fast, cited real-time answers (February 2025). Provider: Perplexity. Released 2025-02-11. Access: API.
Unified ELO 1544 ± 1, rank #465 of 1392 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AI for Education Pedagogy - Technology | 77.36 | Accuracy (%) | 44.5 |
| Gapminder AI Worldview | 78.4 | Correct Rate (%) | 44.1 |
| AI for Education Pedagogy - Social studies | 71.82 | Accuracy (%) | 37.4 |
| AI for Education Pedagogy - Primary | 76.53 | Accuracy (%) | 33.2 |
| BenchmarkList ECI | 99.51 | Capability Index (ECI) | 32.9 |
| Artificial Analysis Intelligence Index | 3.84 | Intelligence Index | 30.9 |
| AA Humanity's Last Exam | 4.88 | Accuracy (%) | 28.5 |
| AI for Education Pedagogy | 70.75 | Accuracy (%) | 28.2 |
| AI for Education Pedagogy - Secondary | 69.34 | Accuracy (%) | 27.3 |
| LM Market Cap LMC Score | 40 | LMC Score (0-100) | 27.3 |
| AI for Education Pedagogy - Science | 66.67 | Accuracy (%) | 23.7 |
| AI for Education Pedagogy - Maths | 64.29 | Accuracy (%) | 22.7 |
Interactive version: theaggregate.ai/model?slug=sonar · How It Works · Data refreshed daily, snapshot 2026-09-05.