Sonar Reasoning Pro
Perplexity’s Sonar Reasoning Pro is a reasoning model with 128K context, enhanced Chain-of-Thought, real-time web retrieval, and no customer-data training.
Model Overview
Capabilities, design details, and architectural traits
Sonar Reasoning Pro - Advanced Reasoning with Enhanced Multi-Step Analysis
Sonar Reasoning Pro is Perplexity’s advanced reasoning model for complex multi-step problem solving. Perplexity describes it as a high-performance reasoning model with enhanced information retrieval and Chain-of-Thought reasoning.
Core Design
Perplexity says the model is built for complex problem-solving that combines advanced reasoning with enhanced information retrieval. The documentation explicitly frames it as a reasoning model rather than a search-only model.
Operating Pattern
The docs show that Sonar Reasoning Pro outputs a thinking section containing reasoning tokens, followed by a valid JSON object. Perplexity also notes that the response_format parameter does not remove those reasoning tokens, so custom parsing might be needed.
Benchmark Performance
Independent evaluations · Artificial Analysis
Compare Models Side-by-Side
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | Perplexity | Anthropic | Anthropic |
| Release Date | January 28, 2025 | September 22, 2026 | September 28, 2026 |
| Knowledge Cutoff | - | - | Jun 2026 |
| Context & Limits | |||
| Context Window | 128K | 1M Best Context Window | 1M Best Context Window |
| Pricing (per 1M tokens) | |||
| Input Pricing | Free Best Input Pricing | $4 | $2 |
| Output Pricing | Free Best Output Pricing | $20 | $10 |
| Modalities | |||
| Inputs | textimage | textimagefile | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 11.8 | 57.6 Best Intelligence Index | 56.0 |
| Coding Index | - | - | - |
| Agentic Index | - | - | - |
Humanity's Last Exam
Extremely difficult logical reasoning and knowledge.
Long Context Reasoning
Logical reasoning over long context windows.
SciCode Benchmark
Scientific coding and mathematical modeling.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
Explore more from Perplexity
Other models by Perplexity
Top AI Models
Leading alternatives by intelligence score
