AI model · unbiased
Pareto 26.10 Preview
Model ID: unbiased/pareto-26.10-preview
- Input
- $0.80 per 1M tokens
- Output
- $3.20 per 1M tokens
- Context
- 1M tokens at once
- Max output
- 131K tokens per reply
- Released
- 1 Oct 2026 unbiased
What it can do
- NoReasoning
- YesTool calling
- YesImage input
- NoOpen weights
- NoFree to use
Takes text, image; gives text.
How its listing describes it
Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks. This is a preview of the...
Sources
- Pareto 26.10 Preview on OpenRouter — read 2 Oct 2026