Perplexity
Perplexity is an AI model provider.Tokenando tracks 7 Perplexity models, with input pricing from $1.000/M and an average blended cost of $4.143/M. Its flagship model is Sonar Pro.
Sonar series: LLMs grounded in live web search. Built for citation-heavy answers rather than pure generation.
MODELS TRACKED
7
3 categories
FLAGSHIP
Sonar Pro
text+image->text
MIN INPUT
$1.000/M
cheapest model in family
AVG BLENDED
$4.143/M
across 7 priced models
MAX CONTEXT
200,000
largest window in family
Reasoning
1 model
Multimodal
4 models
Sonar Proprofile
text+image->text · 200,000 ctx
in $3.000/Mout $15.000/M
tokenizer: Other · cron:openrouter
Sonarprofile
text+image->text · 127,072 ctx
in $1.000/Mout $1.000/M
tokenizer: Other · cron:openrouter
Sonar Reasoning Proprofile
text+image->text · 128,000 ctx
in $2.000/Mout $8.000/M
tokenizer: Other · cron:openrouter
Sonar Pro Search
text+image->text · 200,000 ctx
in $3.000/Mout $15.000/M
tokenizer: Other · cron:openrouter
Efficient
2 models
Frequently Asked Questions
How many models does Perplexity offer?
Tokenando tracks 7 Perplexity models.
How much do Perplexity models cost?
Perplexity model input pricing starts at $1.000 per million tokens, with an average blended cost of $4.143 per million across the 7 priced models we track.
What is Perplexity's flagship model?
Perplexity's flagship model is Sonar Pro. It is the highest-tier Perplexity model we track, with input pricing of $3.000 per million tokens.
What model categories does Perplexity cover?
Perplexity covers 3 categories: reasoning, multimodal and efficient.