Sonar Pro — benchmark results

Perplexity's higher-tier search-grounded API model for deeper retrieval with citations; the Sonar line builds on Llama 3.3 70B (2025). Provider: Perplexity. Released 2025-02-01. Access: API.

Unified ELO 1556 ± 40, rank #607 of 1776 rated models, from 18 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MedBrowseComp 12126.12Accuracy (%)83.3
DeepResearch Bench - Citation Accuracy79.72Score (%)75
AA Humanity's Last Exam7.94Accuracy (%)54.9
AA MMLU-Pro75.52Accuracy (%)50.6
AA MATH-50074.47Accuracy (%)37.1
SnakeBench19TrueSkill Rating35.6
AA GPQA Diamond57.78Accuracy (%)35.4
DeepResearch Bench - Effective Citations16.75Effective Citations33.3
MedBrowseComp 5018Accuracy (%)33.3
Artificial Analysis Intelligence Index9.27Intelligence Index32.1
AA LiveCodeBench27.51Pass@1 (%)30
DeepResearchBench37.76Average Score25.7

Interactive version: theaggregate.ai/model?slug=sonar-pro · How the rankings work · Data refreshed daily, snapshot 2026-07-22.