ModelsCompareBest forBenchmarksStatusPricingAPI

Sonar Deep Research

Perplexity·Operational·Released 2025-02-14text
Context128k
Max output-
Input / 1M$2
Output / 1M$8
Cached in-
Benchmarks6

Benchmarks

Cited public sources · rank is exact-benchmark, apples-to-apples

DRACO Breadth & Depth (Opus 4.6 base)73.1%
DRACO Citation Quality (Opus 4.6 base)64.6%
DRACO Factual Accuracy (Opus 4.6 base)67.9%
DRACO Normalized Score (Opus 4.6 base)70.5%
DRACO Pass Rate (Opus 4.6 base)72.8%
DRACO Presentation Quality (Opus 4.6 base)90.3%

Access Sonar Deep Research and every other model through one endpoint with automatic failover: Respan gateway.