Skip to content
llmwaves

Sonar Reasoning Pro

Perplexity · released Mar 7, 2025

ProprietaryReasoningVision

Note: Sonar Pro pricing includes Perplexity search pricing. See [details here](https://docs.perplexity.ai/guides/pricing#detailed-pricing-breakdown-for-sonar-reasoning-pro-and-sonar-pro) Sonar Reasoning Pro is a premier reasoning model powered by DeepSeek R1 with Chain of Thought (CoT). Designed for...

Specification

Context window
128K
Max output
Knowledge cutoff
Not stated
Parameters
Undisclosed
Licence
Proprietary
Serving providers
1
Moderated
No
Uptime
100.0%

Intelligence

18.0

55th percentile

Coding

Coding Index

Agentic

Agentic Index

Output speed

14 t/s

Median across providers

Latency

26.49s

Time to first token

Cost per task

$0.74

Estimated

Benchmarks

Where the score comes from

The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.

Evaluation scores

Percentage correct · higher is better

  • AIME 2025
    79.0%

An evaluation missing from this list was not run for this model — it is not a zero.

View as table
EvaluationScore
AIME 202579.0%

Against its peers

Intelligence Index · this model highlighted, nearest peers in grey

Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.

View as table
ModelIntelligence
GLM 4.519.7
Kimi K2 071119.7
GPT-4.119.6
o3 Mini19.2
o1-pro19.1
Trinity Large Thinking18.6
Sonar Reasoning Pro18.0
GLM 4.5 Air16.7

Percentile among all indexed models

Intelligence55th

Pricing

What it costs to run

List prices per million tokens, plus what one representative task works out to.

List price

Input / 1M tokens
$2
Output / 1M tokens
$8
Cached input / 1M
Not offered
Blended 3:1
$3.5

One task, estimated

$0.74

Input tokens
50,000
Output tokens
80,000
Profile
Reasoning

Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).

Serving providers

Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.