Sonar Pro: pricing, context window and benchmarks
Sonar Pro is a multimodal model from Perplexity. It offers a 65,536-token context window, costs $11.25 per million input tokens and $46.14 per million output tokens, and averages 77.1 across the index's benchmark suite.
Pricing
Sonar Pro costs $11.25 per million input tokens and $46.14 per million output tokens. A blended 3:1 input-to-output mix works out to about $19.97 per million tokens. Cached input is listed at $1.49 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $11.25 | per 1M tokens |
| Output | $46.14 | per 1M tokens |
| Cached input | $1.49 | per 1M tokens |
| Blended 3:1 | $19.97 | estimate per 1M tokens |
| 1M in + 1M out | $57.39 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 81.1 | 146 | 76th | |
| GPQA Diamond | 69.9 | 269 | 55th | |
| SWE-bench Verified | 67.4 | 160 | 73th | |
| HumanEval | 79.0 | 266 | 56th | |
| MATH-500 | 82.3 | 158 | 74th | |
| AIME 2025 | 75.0 | 154 | 74th | |
| LiveCodeBench | 72.4 | 140 | 77th | |
| IFEval | 83.7 | 221 | 63th | |
| TruthfulQA | 80.8 | 163 | 73th | |
| HellaSwag | 85.5 | 229 | 62th | |
| MMMU | 71.5 | 259 | 57th | |
| Arena Elo | 1413.5 | 175 | 71th |
What Sonar Pro is good for
Sonar Pro is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $11.25 per million input tokens, it sits in the mid-range of the index, and the 65,536-token context window is enough for large files and multi-file diffs.
Strengths
- Accepts more than text, including image input where the provider supports it.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Premium input pricing at $11.25 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| DeepSeek-Coder Chat | DeepSeek | 200000 | $4.20 | 77.3 |
| Yi-Lightning | 01.AI | 32768 | $4.22 | 77.2 |
| Sonar Pro | Perplexity | 65536 | $11.25 | 77.1 |
| MiniMax-Text | MiniMax | 32768 | $4.44 | 77.0 |
| Llama 3.3 Chat | Meta AI | 32768 | $4.18 | 76.9 |
| Llama 3.1 Turbo | Together AI | 200000 | $4.17 | 76.8 |
FAQ
How much does Sonar Pro cost?
Sonar Pro is listed at $11.25 per million input tokens and $46.14 per million output tokens. Cached input is $1.49 per million tokens. A blended 3:1 input-to-output mix is about $19.97 per million tokens.
What is the context window of Sonar Pro?
Sonar Pro has a context window of 65,536 tokens and can emit up to 65,536 output tokens per request in this index.
Is Sonar Pro open weights?
No. Sonar Pro is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Sonar Pro score on benchmarks?
Sonar Pro averages 77.1 across the scored benchmarks in this index, ranking #193 of 600 models. Its strongest relative result is Arena Elo at 1413.5.
What is Sonar Pro best for?
Sonar Pro is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $11.25 per million input tokens, it sits in the mid-range of the index, and the 65,536-token context window is enough for large files and multi-file diffs.