
Perplexity Sonar API
perplexity.ai
Verification pendingResearched, not yet routed
Search-grounded answers and a raw web search API. Buying direct: Search API $5 per 1K requests. Listed .
- Capabilities
- 0
- From
- n/a
- Calls, 30 days
- 0
- Ranked first
- 0 of 0
Overview
Researched from the vendor's own pages; every fact links to its source and the day we checked it.
Sonar models with web-grounded answers, an Agent API with priced tools and a raw Search API; usage tiers by cumulative spend.
Company
- Legal name
- Not confirmed
- Headquarters
- Not confirmed
- Founded
- Not confirmed
- Employees
- Not confirmed
- Ownership
- Not confirmed
- Kind
- Not confirmed
- Website
- Not confirmed
API
- Access
- Self-serve API
- Style
- Not confirmed
- Auth
- Bearer token
- Base URL
https://api.perplexity.ai- Features
- Bulk —Async —Webhooks —MCP —Sandbox —
- Rate limit
- Usage tiers 0-5 by cumulative credit purchases
Pricing
- Model
- pay as you go
- List prices
- Search API $5.00 per 1K requests
- web_search tool $0.0025 per invocation
- Currency
- USD
- Free tier
- Not confirmed
- Checked
- Sep 29, 2026
Compliance
- SOC 2
- Not confirmed
- GDPR
- Not confirmed
- CCPA
- Not confirmed
- DPA
- Not confirmed
- Resale / storage
- Unknown until a written agreement says otherwise
Relationships
- Integrations
- Not confirmed
- Alternatives
- None named by the vendor
Sources (2)
- Perplexity API pricingchecked Sep 29, 2026
- Perplexity rate limitschecked Sep 29, 2026
API
The vendor's OpenAPI document as stored in the catalog, its operations, and the rate limits it publishes.
| Method | Path | Summary | Params | Source |
|---|---|---|---|---|
| POST | /v1/sonar | Create Chat Completion | 0 | Spec |
| POST | /search | Search the Web | 0 | Spec |
| POST | /v1/embeddings | Create Embeddings | 0 | Spec |
| POST | /v1/contextualizedembeddings | Create Contextualized Embeddings | 0 | Spec |
| GET | /v1/async/sonar/{api_request} | Get Async Chat Completion | 1 | Spec |
| GET | /v1/async/sonar | List Async Chat Completions | 0 | Spec |
| POST | /v1/async/sonar | Create Async Chat Completion | 0 | Spec |
| POST | /v1/agent | Create Agent Response | 0 | Spec |
| GET | /v1/agent/{id} | Retrieve Agent Response | 1 | Spec |
| GET | /v1/agent/{id}/files | List Agent Response Files | 1 | Spec |
| GET | /v1/agent/{id}/files/{file_id}/content | Download Agent Response File | 2 | Spec |
| POST | /v1/agent/{id}/cancel | Cancel Agent Response | 1 | Spec |
| GET | /v1/models | List Models | 0 | Spec |
| GET | /v1/analytics/computer/usage | Usage Analytics | 7 | Spec |
| GET | /v2/analytics/computer/usage | Usage Analytics per Member | 6 | Spec |
| GET | /v1/skills | List skills | 0 | Spec |
| POST | /v1/skills | Create a skill | 0 | Spec |
| GET | /v1/skills/{skill_id} | Get a skill | 0 | Spec |
| PUT | /v1/skills/{skill_id} | Add a skill revision | 0 | Spec |
| DELETE | /v1/skills/{skill_id} | Delete a skill | 0 | Spec |
| GET | /v1/skills/{skill_id}/revisions | List skill revisions | 0 | Spec |
| GET | /v1/skills/{skill_id}/download | Get a skill download URL | 0 | Spec |
Rate limits
| Limit | Scope | Plan | Endpoint | Burst / concurrency |
|---|---|---|---|---|
| 1 / second | plan | Tier 0 | Agent API | — / — |
| 3 / second | plan | Tier 1 ($50+) | Agent API | — / — |
| 8 / second | plan | Tier 2 ($250+) | Agent API | — / — |
| 17 / second | plan | Tier 3 ($500+) | Agent API | — / — |
| 33 / second | plan | Tier 4 ($1,000+) | Agent API | — / — |
| 33 / second | plan | Tier 5 ($5,000+) | Agent API | — / — |
Headers: Retry-After. On 429: Rate limited or upstream overloaded; 429s are not billed. GridRouter never paces above these and backs off on 429, honoring Retry-After. Source, checked Sep 29, 2026