> ## Documentation Index
> Fetch the complete documentation index at: https://docs.kymaapi.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Sonar Pro

> Perplexity's pro web-search model on Kyma — deeper multi-step search, 200K context, longer cited reports.

## Overview

`sonar-pro` is Perplexity's pro web-search model. Like [`sonar`](/models/sonar), it answers from a live search of the web with citations — but it searches deeper, reasons over more sources, and writes longer, more complete answers. Use it for complex current-events questions and research tasks where a one-line lookup is not enough.

Sonar Pro does its own searching internally. You send a question; it returns a grounded, cited report. It does not support tool calling.

## Specs

| Field              | Value                                                                     |
| ------------------ | ------------------------------------------------------------------------- |
| Model ID           | `sonar-pro`                                                               |
| Best for           | Deep web research, complex current-events questions, longer cited reports |
| Context window     | 200K                                                                      |
| Max output tokens  | 8K                                                                        |
| Input modalities   | Text, image                                                               |
| Output modalities  | Text                                                                      |
| Tool calling       | No                                                                        |
| Structured outputs | No                                                                        |
| Prompt caching     | No                                                                        |
| Speed              | Medium                                                                    |
| Cost band          | Premium                                                                   |
| Release stage      | Stable                                                                    |

## Pricing

Sonar Pro bills two ways on the same request:

* **Tokens** — $4.05 / 1M input, $20.25 / 1M output.
* **Web-search fee** — a flat **\$0.00675 per request**, on top of tokens, for the live search.

Because the search fee is flat, it matters less here than on `sonar` — Sonar Pro's longer, multi-source answers make the output tokens the larger line item. Every response's `usage.cost` already includes the search fee. See [Pricing → Web search](/pricing#web-search).

## Use this when

* The question is complex and current — multi-part research, comparisons across recent sources, "what changed and why."
* You want a longer, well-cited answer rather than a quick fact.
* You need more context room (200K) for a large prompt plus a long grounded response.

## Pick something else when

* A quick cited lookup is all you need: [`sonar`](/models/sonar) is cheaper per request.
* You need function calling / tools: use [`kimi-k2.6`](/models/kimi-k2.6).
* The task does not need fresh web data: a standard model avoids the per-request search fee entirely.

## Example

```python theme={null}
from openai import OpenAI

client = OpenAI(base_url="https://kymaapi.com/v1", api_key="ky-...")

response = client.chat.completions.create(
    model="sonar-pro",
    messages=[{"role": "user", "content": "Compare the pricing and context windows of the three newest open-weight coding models released this quarter, with sources."}]
)

print(response.choices[0].message.content)
```