Sonar Pro Search vs Qwen3.8 Max (0902)
Qwen3.8 Max (0902) is 2.0× cheaper than Sonar Pro Search on a blended basis. Qwen3.8 Max (0902) has the larger context window at 1M tokens.
Side by side
Green marks the better value in each row where a comparison is meaningful.
| Attribute | Sonar Pro Search | Qwen3.8 Max (0902) |
|---|---|---|
| Provider | Perplexity | Qwen |
| Input / 1M tokens | $3.00 | $2.00 |
| Output / 1M tokens | $15.00 | $6.00 |
| Blended / 1M tokens | $6.00 | $3.00 |
| Cached input / 1M | - | $0.250 |
| Context window | 200K | 1M |
| Max output | 8K | 131K |
| Input modalities | text, image | text, image, video |
| Reasoning | Yes | Yes |
| Tool calling | No | Yes |
| Structured output | Yes | Yes |
| Prompt caching | No | Yes |
| Open weights | No | No |
| Knowledge cutoff | - | - |
| Released | October 30, 2025 | September 3, 2026 |
Frequently asked
Is Sonar Pro Search or Qwen3.8 Max (0902) cheaper?
Qwen3.8 Max (0902) is 2.0× cheaper than Sonar Pro Search on a blended basis.
What is the price difference between Sonar Pro Search and Qwen3.8 Max (0902)?
Sonar Pro Search costs $3.00 per million input tokens and $15.00 per million output tokens. Qwen3.8 Max (0902) costs $2.00 and $6.00 respectively.
Which has the larger context window, Sonar Pro Search or Qwen3.8 Max (0902)?
Qwen3.8 Max (0902), at 1M tokens versus 200K.