Sonar Pro
Language Search-grounded Deprecated Full comparison ↗ perplexity/sonar-pro · by Perplexity
· decoder-only-transformer
Search-grounded model. Unlike a standard base model, this offering combines an underlying LLM with live web retrieval and citations — not a directly comparable like-for-like with a non-search-grounded entry. Billed with the token pricing below plus a per-request search-context fee (see Pricing).
Pricing — 1 offering(s)
Input tokens
- $3.00 / 1M tokens (input) Current 2026-07-26 → present
Output tokens
- $15.00 / 1M tokens (output) Current 2026-07-26 → present
Search context fee (billed on top of token pricing, per query)
- low $6.00 / 1k requests
- medium $10.00 / 1k requests
- high $14.00 / 1k requests
Showing the active price and any recorded history. Full pricing history is available via the paid API — see API docs.
See how Sonar Pro fits into a cost-aware routing setup
See how →Capability profile
How Sonar Pro rates across core capability dimensions, with the task-level evidence behind each rating.
Better than base Sonar at complex queries per Perplexity's positioning, but still a search model — for explicit step-by-step reasoning route to Sonar Reasoning Pro. Perplexity's "not ideal for multi-step analyses" caveat still applies relative to the reasoning/research variants.
Native real-time web search with citations, plus follow-up query support. Not a general function-calling model.
200,000 tokens (Perplexity docs / OpenRouter) — the largest in the Sonar family. Max output is reported as ~8,000 completion tokens (OpenRouter, secondary source; not stated on Perplexity's own docs).
Slower and more expensive than base Sonar due to deeper retrieval; still an interactive-latency search model, not a long-running research agent.
$3 / 1M input, $15 / 1M output, plus a $6-$14 per 1,000 requests search-context fee (low/medium/high). Meaningfully pricier than base Sonar; justified only when the extra search depth or 200K context is actually needed.
Ratings are estimated — limited independent data is available for this model.
Operator guidance
Use Sonar Pro over Sonar when you need more search results per query, the 200K context window, or better handling of complex and follow-up queries, and you can absorb the ~3-5x higher effective cost. Use Sonar Reasoning Pro instead when the value is in explicit chain-of-thought analysis rather than retrieval breadth. Use Sonar Deep Research for exhaustive, hundreds-of-sources reports where a multi-minute latency is acceptable.
Use cases
- Cited web answers to more sophisticated or multi-part questions where base Sonar's single-pass retrieval is too shallow
- Research-style questions that still need an interactive-latency answer (not a multi-minute Deep Research report)
- Workloads needing up to 200K tokens of context alongside live search grounding
- Follow-up / conversational search where earlier turns inform later retrieval
Limitations
- Search-grounded only, with a mandatory per-request search-context fee ($6-$14 / 1K requests) on top of token cost
- Perplexity's own guidance still flags the search models (Sonar and Sonar Pro) as not ideal for exhaustive research or detailed-instruction tasks — those route to the reasoning/research variants
- Max output tokens (~8K) and the underlying base model are not published on Perplexity's own docs; the 8K figure is from OpenRouter
- Legacy Sonar Chat Completions surface migrating to the Agent API (support until 2026-09-27, Perplexity changelog)
- Capability ratings are qualitative from Perplexity's positioning; no independent published benchmark for the API model