Sonar API pricing
As of October 21, 2026, Perplexity Sonar costs $1.00 per million input tokens and $1.00 per million output tokens through OpenRouter.
- Input, per million tokens$1.00
- Output, per million tokens$1.00
- Cached input, per millionNo cache price listed
- Context window127K tokens
- Max output114K tokens
- Output speedNot measured
What a request costs
Three fixed token mixes at the list price, without cache or batch discounts. For a whole month, use the calculator.
| Workload | Tokens | One request | 1,000 requests | Fits the context window |
|---|---|---|---|---|
| Chat turn | 1K input + 500 output tokens | $0.0015 | $1.50 | Yes |
| Repository review | 50K input + 3K output tokens | $0.053 | $53.00 | Yes |
| Agent loop | 200K input + 10K output tokens | $0.210 | $210.00 | No, it would need splitting |
| Solved coding task | One SWE-bench Verified attempt, divided by the share that succeed | Not computable | Not computable |
Price history
We read this price every day and have since October 15, 2026. It has not changed yet; any change will be listed here with its date.
- October 15, 2026 First recorded: $1.00 input, $1.00 output
Cheaper models as good as Sonar
Models within two points of Sonar or above it on the Artificial Analysis Intelligence Index, with a lower output price. Cheapest first.
- Ling 3.0 Flash$0.063 Overall 20.1 against 7.7 · 94% cheaper output
- Llama 3.1 8B Instruct$0.08 Overall 6.9 against 7.7 · 92% cheaper output
- Mistral Small 3$0.08 Overall 6.7 against 7.7 · 92% cheaper output
- gpt-oss-20b$0.09 Overall 10.0 against 7.7 · 91% cheaper output
- Qwen3.5-9B$0.13 Overall 13.3 against 7.7 · 87% cheaper output
Sonar pricing questions
- How much does Sonar cost per million tokens?
- $1.00 per million input tokens and $1.00 per million output tokens, as listed on OpenRouter on October 21, 2026.
- Does Sonar support prompt caching?
- OpenRouter lists no cached input price for this model, so assume every input token is billed at the full rate.
- Is there a cheaper model as good as Sonar?
- Ling 3.0 Flash scores 20.1 on the Artificial Analysis Intelligence Index, within two points of Sonar or above it, and costs $0.063 per million output tokens.
- Has the price of Sonar changed?
- Not since we started recording it daily on October 15, 2026.