Perplexity API pricing
Updated
Perplexity lists standard prices for 24 models. The cheapest is Claude Haiku 5.5 at USD 0.10 input and USD 0.50 output per 1M tokens; the dearest is GPT-6 Astra at USD 10 / USD 50. It publishes cached-input prices for all of them; batch, flex or priority pricing is published as a rule rather than per-model prices (see tiers below).
- Models with a standard price
- 24
- Median input / output
- USD 2 / USD 8.75
- Cached-input prices
- 24 of 24
- Cheapest host for
- 0 of 22 shared models
Which models is Perplexity the cheapest provider for?
None outright. Of the 22 models that Perplexity shares with other providers, 4 are cheaper elsewhere at the standard tier, and Perplexity ties the lowest price for the other 18, where another provider charges the same.
How does Perplexity price the most widely hosted models?
The 6 models Perplexity sells that the most other providers also sell, with Perplexity's rank among them on the 3:1 blend.
| Model | Perplexity | Rank | Cheapest |
|---|---|---|---|
| GLM 5.3 | USD 1.40 / USD 4.40 | 6 of 24 | DeepInfra USD 0.5625 / USD 2.50 |
| GLM 5.3 Flash | USD 0.15 / USD 0.50 | 3 of 20 | GMI Cloud USD 0.075 / USD 0.25 |
| Kimi K3 | USD 3 / USD 15 | 5 of 18 | Chutes USD 1.85 / USD 9.50 |
| DeepSeek V4 Pro 0813 | USD 1.32 / USD 3.96 | 5 of 15 | Chutes USD 0.85 / USD 2.65 |
| Claude Opus 5.5 | USD 4 / USD 20 | 1 of 6 | Anthropic USD 4 / USD 20 (tie) |
| Claude Sonnet 5.5 | USD 2 / USD 10 | 1 of 6 | Anthropic USD 2 / USD 10 (tie) |
Which model makers does Perplexity sell?
Perplexity prices models from 9 makers. OpenAI leads with 8 models. Standard prices, USD per 1M input / output tokens.
| Maker | Models | Cheapest model | Median input / output |
|---|---|---|---|
| OpenAI | 8 | GPT-6 Luna USD 0.10 / USD 0.50 | USD 2 / USD 11 |
| Anthropic | 4 | Claude Haiku 5.5 USD 0.10 / USD 0.50 | USD 3 / USD 15 |
| 3 | Gemini 4 Hydrogen USD 1.50 / USD 7.50 | USD 2 / USD 10 | |
| xAI | 3 | Grok 4.5 USD 2 / USD 6 | USD 2 / USD 6 |
| Z.ai | 2 | GLM 5.3 Flash USD 0.15 / USD 0.50 | USD 0.775 / USD 2.45 |
| DeepSeek | 1 | DeepSeek V4 Pro 0813 USD 1.32 / USD 3.96 | — |
| Moonshot AI | 1 | Kimi K3 USD 3 / USD 15 | — |
| NVIDIA | 1 | Nemotron 3.5 Super USD 0.20 / USD 0.50 | — |
| Perplexity | 1 | Sonar USD 0.25 / USD 2.50 | — |
Perplexity price list
USD per 1M tokens: 24 prices for 24 models, one row per model with any other service tiers stacked under the standard price. Sources checked 2026-11-15 unless a row says otherwise. "vs maker" compares a standard price with the model maker's own API on the 3:1 blend; a minus means Perplexity is cheaper. Of 22 hosted model prices with a maker price to compare, 0 are below the maker's and 0 above.
| Model | Tier | Input | Cached | Output | Context | vs maker | Source |
|---|---|---|---|---|---|---|---|
| Claude Fable 5.2 | standard | USD 8 | USD 0.20 | USD 40 | not stated | same | source |
| Claude Haiku 5.5 | standard | USD 0.10 | USD 0.01 | USD 0.50 | not stated | same | source |
| Claude Opus 5.5 | standard | USD 4 | USD 0.20 | USD 20 | not stated | same | source |
| Claude Sonnet 5.5 | standard | USD 2 | USD 0.10 | USD 10 | not stated | same | source |
| DeepSeek V4 Pro 0813 | standard | USD 1.32 | USD 0.044 | USD 3.96 | not stated | same | source |
| Gemini 4 Argonpromo | standard | USD 2 | USD 0.10 | USD 10 | not stated | same | source |
| Gemini 4 Carbonpromo | standard | USD 2 | USD 0.10 | USD 10 | not stated | same | source |
| Gemini 4 Hydrogen | standard | USD 1.50 | USD 0.075 | USD 7.50 | not stated | same | source |
| GLM 5.3 | standard | USD 1.40 | USD 0.26 | USD 4.40 | not stated | same | source |
| GLM 5.3 Flash | standard | USD 0.15 | USD 0.03 | USD 0.50 | not stated | same | source |
| GPT-5.5 | standard | USD 5 | USD 0.50 | USD 30 | not stated | same | source |
| GPT-5.6 Luna | standard | USD 0.20 | USD 0.02 | USD 1.20 | not stated | same | source |
| GPT-5.6 Sol | standard | USD 4 | USD 0.40 | USD 20 | not stated | same | source |
| GPT-5.6 Terra | standard | USD 2 | USD 0.20 | USD 12 | not stated | same | source |
| GPT-6 Astra | standard | USD 10 | USD 1 | USD 50 | not stated | same | source |
| GPT-6 Luna | standard | USD 0.10 | USD 0.01 | USD 0.50 | not stated | same | source |
| GPT-6 Sol | standard | USD 2 | USD 0.20 | USD 10 | not stated | same | source |
| GPT-6.1 Sol | standard | USD 2 | USD 0.10 | USD 10 | not stated | same | source |
| Grok 4.5 | standard | USD 2 | USD 0.30 | USD 6 | not stated | same | source |
| Grok 4.6 | standard | USD 2 | USD 0.50 | USD 6 | not stated | same | source |
| Grok 4.7 | standard | USD 2 | USD 0.50 | USD 6 | not stated | same | source |
| Kimi K3 | standard | USD 3 | USD 0.30 | USD 15 | not stated | same | source |
| Nemotron 3.5 Super | standard | USD 0.20 | USD 0.04 | USD 0.50 | 1,048,576 | — | source |
| Sonar | standard | USD 0.25 | USD 0.0625 | USD 2.50 | not stated | own model | source |
- Cached "—" means Perplexity's page shows no cache price for that row; budget cached tokens at the input price.
- Context "not stated": the source does not give it (1 of 24 rows have one).
- Promo: Perplexity's page describes 2 prices as promotional or discounted; they can end without notice. Each model page has the list price where the source gives one.
Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).
What service tiers does Perplexity publish?
Perplexity publishes no per-model batch, flex or priority prices, but states these rules: Agent API service tiers (OpenAI models only): "`flex` uses lower-cost, best-effort capacity at 0.5× the listed token prices. `priority` uses higher-priority processing at 2× the listed token prices." (docs.perplexity.ai/docs/agent-api/models). Checked 2026-11-15.
Cached input: 24 of 24 standard prices include a cache-read price, a median 10% of the input price. What cached input means.
What we recorded about Perplexity's pricing page: Token rates from the Agent API Models page (docs.perplexity.ai). The standalone Sonar Chat Completions models (Sonar Pro, Sonar Reasoning Pro, Sonar Deep Research) are no longer on the pricing page; Perplexity points to "Migrate from Sonar to the Agent API". Only perplexity/sonar keeps a token price. Tool invocations (web_search, fetch_url etc.), Search API, embeddings and the Decisions API are billed separately and excluded. Cache-write rates excluded.
Is Perplexity cheaper than the model makers' own APIs?
Mostly it charges the same as the maker. Of 22 standard prices for models whose maker sells its own API, 0 are below the maker's price, 0 above and 22 the same, on the 3:1 blend. The median difference is zero.
BroadNet Inference and Perplexity
BroadRouter is BroadNet's own site, so BroadNet is never ranked on this page. BroadNet Inference also serves 1 model that Perplexity prices: GLM 5.3 at USD 0.78 / USD 3.46 against Perplexity's USD 1.40 / USD 4.40 (BroadNet is cheaper on the 3:1 blend). BroadNet prices are from its rate card, synced 2026-11-14. BroadNet Inference pricing.
More on Perplexity
- Perplexity's own pricing page (the source; checked 2026-11-15)
- All 41 providers compared · Cheapest LLM APIs right now · Download every price (CSV, JSON)
Questions
How much does the Perplexity API cost per token?
Perplexity's standard prices run from USD 0.10 input and USD 0.50 output per 1M tokens (Claude Haiku 5.5) to USD 10 / USD 50 (GPT-6 Astra). The median across 24 standard prices is USD 2 input and USD 8.75 output per 1M, which is USD 0.000002 per input token. Input, cached input and output are billed separately.
What is the cheapest model on Perplexity?
Claude Haiku 5.5, at USD 0.10 input and USD 0.50 output per 1M tokens (USD 0.20 on a 3:1 input-to-output blend). The cheapest from other makers: GPT-6 Luna (OpenAI) at USD 0.10 / USD 0.50; Gemini 4 Hydrogen (Google) at USD 1.50 / USD 7.50; Grok 4.5 (xAI) at USD 2 / USD 6.
Is Perplexity cheaper than other providers?
It depends on the model. Of 22 models that other providers also price, Perplexity is the cheapest standard price for none; it ties the lowest price, without beating it, for 18; another provider is cheaper for 4; and it is the dearest for none. Its median rank is 1 of a median 6 providers, on a 3:1 input-to-output blend.
Does Perplexity have batch, cached or priority pricing?
Cached input: yes, for 24 of 24 standard prices, at a median 10% of the input price. No batch, flex or priority prices are listed per model, but Perplexity states: Agent API service tiers (OpenAI models only): "`flex` uses lower-cost, best-effort capacity at 0.5× the listed token prices. `priority` uses higher-priority processing at 2× the listed token prices." (docs.perplexity.ai/docs/agent-api/models). Checked 2026-11-15.
How much does a chatbot cost per month on Perplexity?
For 2,000 requests a day, 1,500 input tokens (60% cached), 300 output, over 30 days: USD 13.14 on Claude Haiku 5.5, USD 207.00 on Grok 4.7 (the middle of Perplexity's list) and USD 1,314 on GPT-6 Astra. Try your own numbers.