Google Holds 43% Chance of Leading AI Models by End of 2026

A prediction market gives Google a 43% chance of having the leading artificial intelligence model by the end of 2026, narrowly ahead of Anthropic at 41%, as Gemini 4 Argon currently tops a preliminary text leaderboard.

The market will be settled based on which company owns the model ranked No. 1 at a specified check time. The odds reflect trader expectations and do not establish that Google currently has the best model or will keep the lead through 2026.

The Oct. 2, 2026, leaderboard placed Google’s Gemini 4 Argon High first with a score of 1,533±9 from 4,932 votes. Anthropic’s Claude Opus 5.5 High ranked second at 1,512±9 from 4,552 votes. The leaderboard recorded 8,626,731 votes across 413 models, and the Argon result was marked preliminary.

The rankings measure human preferences in text-to-text tasks, including math, coding and creative writing. They do not provide a universal measure of performance across every AI use case.

A separate prediction market focused on the end of November 2026 gave Google a 56% chance and Anthropic 37%. That market is scheduled to resolve Dec. 1, 2026.

Google announced Gemini 4 Argon on Sept. 30, 2026, and said it was initially rolling out to trusted cyber defenders through its Fairwind Program. Google said the model supports a 1 million-token output limit, with introductory pricing of $2 per million input tokens and $10 per million output tokens.

Google also said Argon scored 77.9% on DeepSWE v1.1 and 51.3% on AutomationBench. Anthropic introduced Claude Opus 5.5 on Sept. 22, 2026, and said it costs 40% less to run than Opus 5.

The current leaderboard has Gemini 4 Argon first, while Anthropic models hold multiple positions near the top. The market remains open through 2026, with the end-of-year outcome to be determined by the specified leaderboard ranking.