Sonar Reasoning Pro

perplexity/sonar-reasoning-pro

Perplexity: Sonar Reasoning Pro is a model from Perplexity that accepts text and image inputs and produces text output. It provides a context window of 128,000 tokens, includes reasoning support, and does not support external tools. The model is not open‑weight, was released on 2024‑01‑01, and is priced at $2 per million input tokens and $8 per million output tokens.

  • reasoning

At a glance

providerperplexity
releasedJan 1, 2024
context128K
input / M$2

Specifications

What the API gives you

Context window
128K tokens
Max output
115K tokens
Input price
$2 / 1M tokens
Output price
$8 / 1M tokens
Cache read
—
Modalities
text, image → text
Knowledge cutoff
—
Released
Jan 1, 2024

Measured evidence

Public benchmark record

LMArena Search Factuality‡
1156 Elo2026-11-09high effort

LMArena data licensed CC-BY-4.0. Aider (Apache-2.0) and SWE-bench (MIT) are open. Design Arena uses blind human preference. ‡ Submitted scaffold plus model result. Compare the harness and effort settings, not only the model name.

Task evidence

How broad is the support?

Each status is exact to this model version and the selected task. Official release claims remain provisional.

  • Best LLM for Grounded RAGProvisional evidence

    19% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from CRAG or SimpleQA Verified (Epoch AI). 5 of 7 configured evaluation feeds currently have results.

  • 11% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LMArena Document. 5 of 6 configured evaluation feeds currently have results.

  • 8% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 6 of 7 configured evaluation feeds currently have results.

Recommendation coverage

Where this model appears

Not currently included in a published recommendation. The task evidence panel above says exactly what is confirmed and what is still missing.

Catalog data via OpenRouter and models.dev. Last checked within the past 6 hours.