Skip to content
llmwaves

Sonar

Perplexity · released Jan 27, 2025

ProprietaryVision

Sonar is lightweight, affordable, fast, and simple to use — now featuring citations and the ability to customize sources. It is designed for companies seeking to integrate lightweight question-and-answer features...

Specification

Context window
127K
Max output
Knowledge cutoff
Not stated
Parameters
Undisclosed
Licence
Proprietary
Serving providers
1
Moderated
No
Uptime
100.0%

Intelligence

9.4

34th percentile

Coding

Coding Index

Agentic

Agentic Index

Output speed

54 t/s

Median across providers

Latency

1.84s

Time to first token

Cost per task

$0.08

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

  • MMLU-Pro
    68.9%
  • AIME 2025
    48.7%
  • GPQA Diamond
    47.1%
  • LiveCodeBench
    29.5%
  • SciCode
    22.9%
  • Humanity's Last Exam
    4.9%

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

View as table
EvaluationScore
MMLU-Pro68.9%
AIME 202548.7%
GPQA Diamond47.1%
LiveCodeBench29.5%
SciCode22.9%
Humanity's Last Exam4.9%

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
Llama 4 Scout10.3
Qwen3 VL 30B A3B Instruct9.9
R1 Distill Llama 70B9.8
GPT-4.1 Nano9.6
Sonar9.4
Qwen2.5 72B Instruct9.4
GPT-4o (2024-08-06)9.4
Sonar Pro9.1

Percentile among all indexed models

Intelligence34th

Pricing

What it costs to run

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

List price

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

One task, estimated

$0.08

Input tokens
50,000
Output tokens
25,000
Profile
Standard

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.