Sulat.com
AI models
$10 off the fastest DeepSeek V4.1 Flash, Kimi K3 and GLM 5.3 from Synthetic
Kilo Gateway logo

Model details

Perplexity: Sonar Pro Search

Perplexity built Sonar Pro Search as an advanced agentic search system designed to move beyond single-query synthesis. Where traditional models process one prompt and deliver a response, this system plans and executes entire research workflows autonomously using tool use. It represents the most capable iteration in the Sonar family, built to handle complex reasoning and analysis tasks that require structured, multi-step approaches. The system accepts both text and image inputs, making it suitable for visually-informed research and document-heavy analysis. This architectural design positions it as a research engine rather than a simple chatbot.

This model powers Perplexity's own Pro Search mode, meaning the capabilities visible on the platform directly reflect this API offering. The system excels at in-depth information synthesis, autonomous workflow execution, and structured research planning. Practical applications include comprehensive research projects, data analysis, detailed report generation, and complex information workflows. Its autonomous multi-step capabilities make it especially valuable for users who need thorough, structured outputs rather than quick single-turn answers.

Kilo Gatewayperplexity/sonar-pro-searchsonar-pro

Quick Info

Powered by
Provider
Kilo Gateway
Model key
perplexity/sonar-pro-search
Release date
Oct 30, 2025
Last updated
Oct 30, 2025
Input modalities
Output modalities
Capabilities

Cost

Input token cost
$3.00
Output token cost
$15.00

Limits

Output tokens
8,000 tokens
Context window
200,000 tokens

Transparent token rates

Compare Perplexity: Sonar Pro Search pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about Perplexity: Sonar Pro Search

Kilo Gateway

CoverageBenchmark

OpenOCR published a dedicated benchmark page for Perplexity: Sonar Pro Search on August 22, 2026, reporting an average OCR score of 98.8% across 11 documents in the corpusocr-shared-11-normalized/v2 set, with an average latency of 4.2 seconds. All 11 per-document scores are marked reproducible, and each score is recomp Per-document breakdowns show strong performance on dense numeric rows (Grocery Receipt 100.0%, Café Receipt 98.6%) and on handwriting (Handwritten Letter 100.0%, Handwritten Notes 100.0%), with the page providing ground-truth character counts, returned character counts, and normalized character-LCS similarity for each.

Videos about Perplexity: Sonar Pro Search

More models around Perplexity: Sonar Pro Search