Sonar Max: pricing, context window and benchmarks
Sonar Max is a multimodal model from Perplexity. It offers a 32,768-token context window, costs $11.60 per million input tokens and $42.81 per million output tokens, and averages 77.8 across the index's benchmark suite.
Pricing
Sonar Max costs $11.60 per million input tokens and $42.81 per million output tokens. A blended 3:1 input-to-output mix works out to about $19.40 per million tokens. Cached input is listed at $1.41 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $11.60 | per 1M tokens |
| Output | $42.81 | per 1M tokens |
| Cached input | $1.41 | per 1M tokens |
| Blended 3:1 | $19.40 | estimate per 1M tokens |
| 1M in + 1M out | $54.41 | 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 | 81.3 | 141 | 77th | |
| GPQA Diamond | 73.7 | 180 | 70th | |
| SWE-bench Verified | 63.0 | 253 | 58th | |
| HumanEval | 82.1 | 186 | 69th | |
| MATH-500 | 83.3 | 143 | 76th | |
| AIME 2025 | 74.9 | 156 | 74th | |
| LiveCodeBench | 72.8 | 133 | 78th | |
| IFEval | 85.2 | 188 | 69th | |
| TruthfulQA | 78.9 | 224 | 63th | |
| HellaSwag | 85.3 | 232 | 61th | |
| MMMU | 75.4 | 156 | 74th | |
| Arena Elo | 1395.2 | 227 | 62th |
What Sonar Max is good for
Sonar Max 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 $11.60 per million input tokens, it sits in the mid-range of the index, and the 32,768-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 $11.60 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| ERNIE 4.0 Reasoning | Baidu | 65536 | $12.20 | 77.8 |
| MiniMax-Text Chat | MiniMax | 128000 | $4.26 | 77.8 |
| Sonar Max | Perplexity | 32768 | $11.60 | 77.8 |
| DeepSeek-Coder Instruct | DeepSeek | 32768 | $4.34 | 77.7 |
| GPT-5.1 Mini | OpenAI | 128000 | $12.19 | 77.7 |
| o4 Mini | OpenAI | 200000 | $11.92 | 77.7 |
FAQ
How much does Sonar Max cost?
Sonar Max is listed at $11.60 per million input tokens and $42.81 per million output tokens. Cached input is $1.41 per million tokens. A blended 3:1 input-to-output mix is about $19.40 per million tokens.
What is the context window of Sonar Max?
Sonar Max has a context window of 32,768 tokens and can emit up to 32,768 output tokens per request in this index.
Is Sonar Max open weights?
No. Sonar Max is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Sonar Max score on benchmarks?
Sonar Max averages 77.8 across the scored benchmarks in this index, ranking #178 of 600 models. Its strongest relative result is Arena Elo at 1395.2.
What is Sonar Max best for?
Sonar Max 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 $11.60 per million input tokens, it sits in the mid-range of the index, and the 32,768-token context window is enough for large files and multi-file diffs.