Perplexity Sonar
PerplexityLarge Language ModelProprietaryAI search model that combines LLM capabilities with real-time web search, returning cited, up-to-date answers. DEPRECATED: the Sonar Chat Completions API is supported only until September 27, 2026 — Perplexity directs developers to the new Agent API. Request fees are now tiered at $5-12 per 1K requests by search context size.
Abilities
Use Cases
Available in Tools
Availability
How to Use
Pros
- Real-time web search built-in — always has current information
- Citations with every answer — verifiable sources you can check
- Excellent for research and fact-checking tasks
- OpenAI-compatible API — easy to integrate into existing pipelines
Cons
- Dependent on web search quality — garbage in, garbage out
- Can occasionally hallucinate citations that look real but are wrong
- Less capable on pure reasoning compared to dedicated reasoning models
- Search costs ($5/1000 searches) add up for high-volume use
- Closed source — no self-hosting option
What to Use It For
Perfect For
Real-time web search ensures answers are based on the latest available data
Every answer includes clickable citations — you can verify claims directly
Good For
Combines web search with synthesis — good for literature review and exploration
Real-time access to news sources enables current market and competitor analysis
Not Recommended
Optimized for search and research, not code generation or debugging
Try instead: Claude Sonnet 5
Search-focused model — not designed for narrative or creative text generation
Try instead: Claude Opus 5
Do Not Use For
The entire value proposition is real-time web search — useless without internet
Try instead: Llama 4 Maverick
Text-only model — cannot create images
Try instead: FLUX.1