Sonar Pro Reasoning: pricing, context window and benchmarks
Sonar Pro Reasoning is a multimodal model from Perplexity. It offers a 128,000-token context window, costs $13.94 per million input tokens and $72.76 per million output tokens, and averages 80.5 across the index's benchmark suite.
Pricing
Sonar Pro Reasoning costs $13.94 per million input tokens and $72.76 per million output tokens. A blended 3:1 input-to-output mix works out to about $28.65 per million tokens. Cached input is listed at $1.87 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $13.94 | per 1M tokens |
| Output | $72.76 | per 1M tokens |
| Cached input | $1.87 | per 1M tokens |
| Blended 3:1 | $28.65 | estimate per 1M tokens |
| 1M in + 1M out | $86.70 | 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 | 82.6 | 112 | 81th | |
| GPQA Diamond | 76.2 | 116 | 81th | |
| SWE-bench Verified | 71.7 | 82 | 86th | |
| HumanEval | 82.7 | 166 | 72th | |
| MATH-500 | 85.1 | 108 | 82th | |
| AIME 2025 | 73.7 | 194 | 68th | |
| LiveCodeBench | 75.6 | 90 | 85th | |
| IFEval | 87.1 | 131 | 78th | |
| TruthfulQA | 84.4 | 77 | 87th | |
| HellaSwag | 89.9 | 89 | 85th | |
| MMMU | 76.1 | 137 | 77th | |
| Arena Elo | 1439.7 | 97 | 84th |
What Sonar Pro Reasoning is good for
Sonar Pro Reasoning 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 $13.94 per million input tokens, it sits in the mid-range of the index, and the 128,000-token context window is enough for large files and multi-file diffs.
Strengths
- Large 128,000-token context window for long documents and agent traces.
- Accepts more than text, including image input where the provider supports it.
- Reasoning-oriented profile, useful for multi-step problems and structured analysis.
- 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 $13.94 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| ERNIE 4.5 Max | Baidu | 128000 | $14.41 | 80.6 |
| MiniMax-M2 | MiniMax | 200000 | $5.19 | 80.6 |
| ERNIE 4.5 Pro | Baidu | 1000000 | $13.89 | 80.5 |
| Kimi K1.5 Instruct | Moonshot AI | 200000 | $5.15 | 80.5 |
| Sonar Pro Reasoning | Perplexity | 128000 | $13.94 | 80.5 |
| GPT-5.2 Mini | OpenAI | 200000 | $13.34 | 80.4 |
FAQ
How much does Sonar Pro Reasoning cost?
Sonar Pro Reasoning is listed at $13.94 per million input tokens and $72.76 per million output tokens. Cached input is $1.87 per million tokens. A blended 3:1 input-to-output mix is about $28.65 per million tokens.
What is the context window of Sonar Pro Reasoning?
Sonar Pro Reasoning has a context window of 128,000 tokens and can emit up to 32,768 output tokens per request in this index.
Is Sonar Pro Reasoning open weights?
No. Sonar Pro Reasoning is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Sonar Pro Reasoning score on benchmarks?
Sonar Pro Reasoning averages 80.5 across the scored benchmarks in this index, ranking #108 of 600 models. Its strongest relative result is Arena Elo at 1439.7.
What is Sonar Pro Reasoning best for?
Sonar Pro Reasoning 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 $13.94 per million input tokens, it sits in the mid-range of the index, and the 128,000-token context window is enough for large files and multi-file diffs.