Sonar Pro Mini: pricing, context window and benchmarks
Sonar Pro Mini is a mid-range model from Perplexity. It offers a 8,192-token context window, costs $8.02 per million input tokens and $28.92 per million output tokens, and averages 70.3 across the index's benchmark suite.
Pricing
Sonar Pro Mini costs $8.02 per million input tokens and $28.92 per million output tokens. A blended 3:1 input-to-output mix works out to about $13.25 per million tokens. Cached input is listed at $0.84 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $8.02 | per 1M tokens |
| Output | $28.92 | per 1M tokens |
| Cached input | $0.84 | per 1M tokens |
| Blended 3:1 | $13.25 | estimate per 1M tokens |
| 1M in + 1M out | $36.94 | 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 | 73.8 | 358 | 40th | |
| GPQA Diamond | 68.7 | 303 | 50th | |
| SWE-bench Verified | 56.7 | 371 | 38th | |
| HumanEval | 77.4 | 301 | 50th | |
| MATH-500 | 74.8 | 333 | 45th | |
| AIME 2025 | 64.0 | 356 | 41th | |
| LiveCodeBench | 59.3 | 375 | 38th | |
| IFEval | 77.9 | 369 | 39th | |
| TruthfulQA | 73.1 | 367 | 39th | |
| HellaSwag | 80.7 | 376 | 37th | |
| MMMU | 67.3 | 356 | 41th | |
| Arena Elo | 1354.4 | 356 | 41th |
What Sonar Pro Mini is good for
Sonar Pro Mini is a general-purpose model in the index. It handles text generation, summarization and question answering, and it can be a sensible default when you need a balance of quality, price and context. At $8.02 per million input tokens with a 8,192-token window, check its benchmark profile against the specific task before committing.
Strengths
- Accepts more than text, including image input where the provider supports it.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Premium input pricing at $8.02 per million tokens.
- Context window of 8,192 tokens is small for long-document work.
- Average benchmark score is below the index median; test it on your own task.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| GPT-4.1 Nano | OpenAI | 32768 | $7.42 | 70.5 |
| Firesearch Pro | Fireworks AI | 32768 | $7.67 | 70.4 |
| Llama 3 Novita Base | Novita AI | 8192 | $3.06 | 70.3 |
| Sonar Pro Mini | Perplexity | 8192 | $8.02 | 70.3 |
| Yi-Large Base | 01.AI | 16384 | $2.78 | 70.3 |
| Auto | OpenRouter | 32768 | $7.53 | 70.2 |
FAQ
How much does Sonar Pro Mini cost?
Sonar Pro Mini is listed at $8.02 per million input tokens and $28.92 per million output tokens. Cached input is $0.84 per million tokens. A blended 3:1 input-to-output mix is about $13.25 per million tokens.
What is the context window of Sonar Pro Mini?
Sonar Pro Mini has a context window of 8,192 tokens and can emit up to 8,192 output tokens per request in this index.
Is Sonar Pro Mini open weights?
No. Sonar Pro Mini is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Sonar Pro Mini score on benchmarks?
Sonar Pro Mini averages 70.3 across the scored benchmarks in this index, ranking #360 of 600 models. Its strongest relative result is Arena Elo at 1354.4.
What is Sonar Pro Mini best for?
Sonar Pro Mini is a general-purpose model in the index. It handles text generation, summarization and question answering, and it can be a sensible default when you need a balance of quality, price and context. At $8.02 per million input tokens with a 8,192-token window, check its benchmark profile against the specific task before committing.