Sonar Reasoning
sonar-reasoningPricing
List prices at Perplexity, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $1.00 | $5.00 | |
| Web search $8.00 per 1,000 searches | $8.00 per 1,000 searches |
What it costs a month
At Perplexity list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out$675
- RAG over documents2,000 requests a day, 8K in / 0.6K out$660
- Coding agent1,000 requests a day, 40K in / 2K out$1,500
- Overnight batch20,000 requests a day, 4K in / 0.4K out$3,600
Where to run it
2 hosts serve Sonar Reasoning. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
PerplexityGateway sonar-reasoning | $1.00 / $5.00 | sonar-reasoning | |
Vercel AI GatewayGateway sonar-reasoning | $1.00 / $5.00 | sonar-reasoning |
Capabilities
Core
- Streaming
- Function calling
- JSON mode
- Structured outputs
Multimodal
- Vision
- Image input
- Audio
- Image generation
- Video
Advanced
- Reasoning
- Tool calling
- MCP
- Prompt caching
- Batch API
- Embeddings
- Takes in
- Text
- Produces
- Text
- Features
- Built-in web search
Cheaper alternatives
Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.
Same model, cheaper host
No host sells it for less than Perplexity.
Smaller Perplexity models
No cheaper Perplexity model keeps its key capabilities.
Other frontier labs
- 16% less
- 22% less
- 25% less
- 90% less
Open-weight picks
- 14% less
- 57% less
- 59% less
- 60% less
Call it
Using Perplexity's API.
from litellm import completion
response = completion(
model="perplexity/sonar-reasoning",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does Sonar Reasoning cost?
On Perplexity, Sonar Reasoning costs $1.00 per million input tokens and $5.00 per million output tokens.
Which APIs offer Sonar Reasoning?
Sonar Reasoning runs on 2 APIs: Perplexity and Vercel AI Gateway.
What is Sonar Reasoning's context window?
128K tokens, with up to 128K tokens in one reply.
Is Sonar Reasoning being retired?
No retirement date is scheduled for it on any API we track.
What are cheaper alternatives to Sonar Reasoning?
Models that cost less and keep most of what it can do: GPT-5.4 Mini on OpenAI (16% less), Grok 4.20 Multi Agent on xAI (22% less) and Inkling on DeepInfra (14% less). Savings are on a mix of 3 input tokens to 1 output token.
What is Sonar Reasoning's API model id?
Use sonar-reasoning with Perplexity.
Sources
Raw catalog entry for sonar-reasoning on Perplexity
{
"input_cost_per_token": 0.000001,
"litellm_provider": "perplexity",
"max_input_tokens": 128000,
"max_output_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 0.000005,
"search_context_cost_per_query": {
"search_context_size_high": 0.014,
"search_context_size_low": 0.005,
"search_context_size_medium": 0.008
},
"supports_reasoning": true,
"supports_web_search": true
}