Perplexity Agent API Advanced Deep Research Preset

The most powerful Perplexity Agent API preset using Claude Opus 4.6 for institutional-grade research with maximum depth, 10 reasoning steps, cross-source validation, and 10K token output capacity.

Overview

The advanced-deep-research preset represents the highest capability tier of Perplexity's Agent API. It combines Anthropic's Claude Opus 4.6 with Perplexity's search infrastructure for the most thorough research possible.

Configuration

• Model: anthropic/perplexity-opus-4-6
• Max Output Tokens: 10,000
• Max Tokens Per Page: 4,000
• Max Steps: 10
• Tools: web_search, fetch_url
• Prompt Token Count: ~3,500

Capabilities

• Iterative research up to 10 steps per query
• Cross-validation across multiple independent sources
• Length calibration by query complexity
• Primary source prioritization over secondary reporting
• Institutional-grade depth and rigor
• Full inline citation support

Usage

from perplexity import Perplexity

client = Perplexity()
response = client.responses.create(
    preset="advanced-deep-research",
    input="Detailed regulatory impact assessment of the EU AI Act on enterprise AI deployment"
)

When to Use

Use for legal research, academic literature reviews, competitive intelligence reports, regulatory analysis, and any query requiring maximum thoroughness and source quality.