2026-10-04 02:55 UTC · perplexity/api-pricing · live page
Summary by fallback. It may be wrong; the diff below is what actually changed. Raw: 000cadb7f0-perplexity-api-pricing.diff · commit 000cadb7f0
diff --git a/snapshots/perplexity/api-pricing.md b/snapshots/perplexity/api-pricing.mdindex e39e047..ba36501 100644--- a/snapshots/perplexity/api-pricing.md+++ b/snapshots/perplexity/api-pricing.md@@ -11,5 +11,5 @@ Router API usage is billed per token at each model's published rates — there a ## Agent API Pricing -The Agent API provides access to third-party models from providers including OpenAI, Anthropic, Google, xAI, Z.AI, Moonshot AI, and NVIDIA with transparent, token-based pricing at direct provider rates with no markup.+The Agent API provides access to third-party models from providers including OpenAI, Anthropic, Google, xAI, Z.AI, Moonshot AI, and NVIDIA with transparent, token-based pricing at each model's published rates. ### Model Pricing@@ -27,5 +27,7 @@ When using tools with the Agent API: | Tool | Price | Description | |---|---|---|-| `web_search` | $0.0025 per invocation | Performs web searches to retrieve current information |+| `web_search` (standard) | $0.0025 per invocation | Standard web search with `search_type: "web"`. $2.50 per 1,000 invocations |+| `web_search` (Fast Search) | $0.001 per invocation | The same tool with `search_type: "fast"`. $1.00 per 1,000 invocations. See Fast Search |+| `image_search` | $0.0025 per successful invocation | Finds structured image results with domain, format, and safe-search filters. $2.50 per 1,000 successful invocations | | `fetch_url` | $0.0005 per invocation | Fetches and extracts content from specific URLs | | `people_search` | $0.005 per invocation | Looks up professionals, employees, and people. $5 per 1,000 tool invocations |@@ -40,58 +42,8 @@ Most tool costs are per invocation. `sandbox` is billed per container session |---|---|---| | Search API | $5.00 | Raw web search results with advanced filtering |+| Search API with Fast Search | $1.00 | Lower-latency web search results. Set `search_type: "fast"` | Billing unit: Search API charges for each successful `POST /search` request, not for each query in the request. A successful request containing an array of up to five queries is one billing unit. Invalid requests, rate-limited requests, and upstream failures are not billed. A successful response is billed even when it returns no results. There are no additional token-based charges. -## Sonar API Pricing--Total cost per query = Token costs + Request fee (varies by search context size, applies to Sonar, Sonar Pro, and Sonar Reasoning Pro models only)--- Token Pricing-- Request Pricing-- Pro Search Pricing--## Token Pricing--Token pricing is based on the number of tokens in your request and response.--| Model | Input Tokens ($/1M) | Output Tokens ($/1M) | Citation Tokens ($/1M) | Search Queries ($/1K) | Reasoning Tokens ($/1M) |-|---|---|---|---|---|---|-| Sonar | $1 | $1 | - | - | - |-| Sonar Pro | $3 | $15 | - | - | - |-| Sonar Reasoning Pro | $2 | $8 | - | - | - |-| Sonar Deep Research | $2 | $8 | $2 | $5 | $3 |--## Request Pricing by Search Context Size--Search context determines how much web information is retrieved. Higher context = more comprehensive results. The following table shows the request fee for each model for every 1000 requests.--| Model | Low Context Size | Medium Context Size | High Context Size |-|---|---|---|---|-| Sonar | $5 | $8 | $12 |-| Sonar Pro | $6 | $10 | $14 |-| Sonar Reasoning Pro | $6 | $10 | $14 |--- Low: (default) fastest, cheapest-- Medium: Balanced cost/quality-- High: Maximum search depth, best for research--Learn more about search context →--## Pro Search Pricing (Pro Search for Sonar Pro)--Pro Search enhances Sonar Pro with automated tool usage and multi-step reasoning. When enabled, the model can perform multiple web searches and fetch URL content to answer complex queries. Learn more about Pro Search here.--Pro Search requires `stream: true` and is enabled via the `search_type` parameter in `web_search_options`.--### Search Type Options--| Search Type | Description | Request Fee (per 1K) |-|---|---|---|-| `fast` | (default) Standard Sonar Pro behavior | $6 / $10 / $14 |-| `pro` | Multi-step tool usage for complex queries | $14 / $18 / $22 |-| `auto` | Automatic classification based on query complexity | Varies by classification |--Request fees vary by search context size (Low / Medium / High). Token pricing remains the same as standard Sonar Pro ($3 per 1M input, $15 per 1M output).- ## Embeddings API Pricing @@ -116,4 +68,18 @@ Generate high-quality text embeddings for semantic search, retrieval-augmented g Learn how to use the Embeddings API for semantic search, RAG, and more. +## Decisions API Pricing++The Decisions API answers yes/no, multiple-choice, and scored questions about text and images with calibrated probabilities.++| Model | Price ($/1M input tokens) | Output tokens | Per-request fee |+|---|---|---|---|+| `pplx-decider-v1-27b` | $0.04 | Free | None |++Billing unit: you pay for input tokens only. The `state`, every question, and any images count as input; the response reports the exact number as `usage.input_tokens`, so you can compute the cost of a request from the response you already receive. A request with 600 input tokens costs $0.000024.++## View Decisions API Documentation++Learn how to ask structured questions and read the probabilities the Decisions API returns.+ Token and Cost Glossary @@ -138,28 +104,11 @@ The number of tokens in the API's response. This includes: Example: "The weather in New York is currently sunny with a temperature of 72°F." = ~15 output tokens -### Citation Tokens--Tokens used specifically for generating search results and references in responses. Only applies to Sonar Deep Research model.Example: Including source links, reference numbers, and bibliographic information- ### Search Context Size vs Context Window Search context size is not the same as the context window. -- Search context size: How much web information is retrieved during search (affects request pricing)+- Search context size: How much web information is retrieved during search - Context window: Maximum tokens the model can process in one request (affects token limits) -### Search Queries--The number of individual searches conducted by Sonar Deep Research during query processing. This is separate from your initial user query.--- The model automatically determines how many searches are needed-- You cannot control the exact number of search queries-- The `reasoning_effort` parameter influences the number of searches performed-- Only applies to Sonar Deep Research model--### Reasoning Tokens--Tokens used for step-by-step logical reasoning and problem-solving. Only applies to Sonar Deep Research model.Example: Breaking down a complex math problem into sequential steps with explanations- Token Calculation: 1 token ≈ 4 characters in English text. The exact count may vary based on language and content complexity. @@ -168,12 +117,12 @@ Token Calculation: 1 token ≈ 4 characters in English text. The exact count may ## Agent API Web Search -`openai/gpt-5.2` • 500 input + 200 output tokens • 1 web search+`openai/gpt-5.6-terra` • 500 input + 200 output tokens • 1 web search | Component | Cost | |---|---|-| Input tokens | $0.000875 |-| Output tokens | $0.0028 |+| Input tokens | $0.001 |+| Output tokens | $0.0024 | | `web_search` | $0.0025 |-| Total | $0.006175 |+| Total | $0.0059 | ## Agent API Research Preset