Sonar

Provider: Perplexity US Official API price

Input price
$1.00
/ 1M tokens
Output price
$1.00
/ 1M tokens
Context
127K tokens
Max output 114K tokens

Overview

Perplexity Sonar is a cost-optimized model from Perplexity (US), designed for general-purpose workloads that require a substantial context window and multimodal vision capabilities. It targets developers seeking highly competitive pricing for both input and output token processing.

Strengths

  • Extremely competitive $1/M token pricing for both input and output.
  • Supports multimodal vision capabilities for diverse applications.
  • Features a substantial context window of 127072 tokens.
  • Provides balanced input and output token costs, simplifying expense prediction.
  • Cost-effective option for large-scale text and image processing.

Best use cases

  • Analyzing extensive documents combined with visual data efficiently.
  • Building cost-optimized multimodal applications requiring image understanding.
  • Developing tools for summarizing long articles and related images.
  • Implementing applications where consistent low I/O token costs are crucial.
  • Processing large volumes of text and visual information at scale.

Pricing position

Perplexity Sonar is the most affordable model from Perplexity, priced at an impressive $1/M for both input and output tokens, significantly less than siblings like Sonar Reasoning Pro (in $2/M, out $8/M) and Sonar Pro (in $3/M, out $15/M). Compared to peers, its $1/M output price is exceptionally competitive, often beating Grok Build 0.1 ($2/M), GLM 5.1 ($3.08/M), Claude Haiku 4.5 ($5/M), and various Qwen models (up to $6.24/M) at the same input cost.

Caveats

Perplexity Sonar does not offer a free tier, requiring users to pay for all usage from the start. Its maximum output length of 4000 tokens may be insufficient for tasks requiring extremely long generations. For workloads demanding higher reasoning or more advanced capabilities beyond multimodal processing, users might need to consider the more expensive Sonar Reasoning Pro model.

Notes drafted by an LLM from the figures on this page (2026-05-21) and rewritten when the price changes.

Key facts

Price basis
Official API price
Max output
114K tokens
Listed on OpenRouter
2025-01-27
Tracked since
2026-05-07
Last updated
2026-11-10

Features

Image input

Monthly cost estimate

Scale Input (M) Output (M) Monthly
0.1M (hobby) 0.1 0.05 $0.15
1M (small app) 1 0.5 $1.50
10M (mid scale) 10 5 $15.00
100M (large) 100 50 $150.00

※ Cache and batch discounts not factored in. Detailed calculator

Official resources

Other models from Perplexity

Similar-priced models

Source: OpenRouter API (perplexity/sonar)