Unbiased launched Pareto on Venice on November 12, 2026, with a 1,048,576-token context window, image input, and anonymous access without identity tracking. Pareto is a composite system that runs multiple frontier and open models in parallel, then synthesizes their outputs instead of assigning each question to a single model. The 26.10 Preview version, released on October 1, 2026, quadrupled the context window from 262,000 tokens to more than one million. It is priced at approximately $0.80 per million input tokens and $3.20 per million output tokens, making output four times as expensive. The launch targets developers and researchers who analyze long documents or large codebases while benefiting from local conversation storage and minimal data retention on Venice.
Source: Read the original article


