Perplexity: Sonar is Perplexity's model. Its measured tokenizer sits with the Llama3 group, which is evidence that it builds on the Llama3 tokenizer. Tokenizer reuse is normal engineering, open vocabularies travel between labs, and this says nothing against Perplexity's authorship of the model itself.
| Signal | Result | Reference |
|---|---|---|
| Tokenizer signature | Exact match with 7 other models | tk_6588ed5b |
| API surface | No other entry shares this exact contract | api_d6eda9b6 |
| Context and output | 127,072 context · 114,364 max output | declared |
| Reasoning contract | None declared | declared |
| Serving providers observed | Perplexity | measured 3 September 2026 |
Click a row to open the full comparison.
An x marks a row excluded as corrupt (caching or provider interference detected during measurement). Overhead subtracted: 0 prompt tokens.
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...
Supported parameters
frequency_penaltymax_tokenspresence_penaltytemperaturetop_ktop_pweb_search_options
defaults: {}
Cite this page as the evidence record for Perplexity: Sonar: the URL is stable, measurements are dated, and revisions append to the history above. Method: tokenizer fingerprinting, v1, 50 strings.
