ModelPriceWatch.com
Last scan 2026-11-22 Models tracked 322 Providers 42 Cheapest paid pplx-decider-flash-v1.3 $0.010/Mtok in Every price links to its source

Sonar (Agent API)

by Perplexity

Current budget cheap tier
Availability: Sold by Perplexity through its Agent API under the model name perplexity/sonar. Perplexity's model table prints no context window and no release date for it, so neither is published here. OpenRouter's Perplexity Sonar listing is still the old Chat Completions model at $1/$1, not this one.

Today's price · per 1M tokens

Input

$0.250

per 1M tokens

Output

$2.50

per 1M tokens

Blended

$0.813

blended $/1M — 3:1 weighted input:output

Cached input

$0.063

25% of input — prompt caching

How this price is scoped: First-party list price. Perplexity's Agent API model table prints $0.25 input, $2.50 output and $0.0625 cache read per 1M tokens for perplexity/sonar, with default processing only (no flex or priority tier). This is a different product from the retired Sonar Chat Completions model, which cost $1/$1 — not a price cut on it. Perplexity's pricing page bills tools separately from tokens, per call: web search $0.0025 per search ($0.001 with Fast Search), URL fetch $0.0005. A request that searches costs its tokens plus its searches.

Source: official Perplexity pricing · read MODELPRICEWATCH.COM · 2026-11-22

Price receipt

We read Perplexity's own pricing page on and found perplexity/sonar listed with a price on the same row — that read is where the number above comes from. We keep the page text we read; its content hash is 9c8a340f4d.

Overview

Perplexity's own grounded search model, sold on its Agent API — the API Perplexity moved Sonar users to when it ended Sonar Chat Completions support on 2026-09-27. $0.25/$2.50 per 1M tokens; cached input $0.0625. Web searches are billed on top of tokens as a separate tool charge: $2.50 per 1,000 standard searches, $1.00 per 1,000 Fast Search calls.

Capabilities

struck through = not supported
Input 1/4
Text ✓ Image Audio Video
Output
Text ✓
Features 2/9
Prompt caching ✓ Reasoning Coding Fast inference Long context Open weights Multimodal Web search ✓ Realtime

Benchmark performance

No benchmark scores are shown for this model. Perplexity publishes no benchmark for Sonar on its Agent API. Artificial Analysis scores the Sonar Chat Completions endpoint Perplexity ended on 2026-09-27, and Perplexity does not say it is the same model, so those scores are not reused. Vals lists no score (checked 2026-11-21).

Specifications

Provider
Perplexity
Context window
— tokens
Modality
text
Parameters
Proprietary
Open source
No — proprietary
Released
—
Status
Current
Last updated
Tags
budgetsearchcaching

Availability verified: — listed on Perplexity's own page