BenchLeader
PerplexityReasoning modelAuto-detected

Perplexity Sonar Deep Research

Not yet ranked: too few independent results so far. Released 1 Feb 2025.

Blended price
$3.50/M
$2.00 in · $8.00 out
Output speed
First answer
Context
128k

No independent quality benchmark results yet — only pricing and metadata.

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.

WorkloadTokens in / outCostWith cachingTime
Chat reply400 / 300$0.0032
Summarise a 30-page report12,000 / 600$0.029
Code edit6,000 / 1,500$0.024
Agentic coding session60,000 / 4,000$0.152
Structured extraction2,000 / 200$0.0056

See also

Data as of 9 Sept 2026. Compare with another model.