Gemini 4 Argon

Google's flagship reasoning model, generally available since October 13, 2026. Standard API rates are $2/$0.10/$10 per M input/cached-input/output tokens through November 30, then $4/$0.20/$20 from December 1. Batch and Flex are half of Standard; Priority is 1.8x. Implicit caching only; explicit caching is unsupported. The default synchronous output limit is 65,536 tokens; background long decode supports up to 1,048,576 output tokens, limited by the shared context window.

Google Gemini API release notes, checked 2026-10-14, mark Gemini 4 Argon (`gemini-4-argon`) generally available on October 13. Google's Gemini 4 Argon model page, checked 2026-10-14, gives a 1,048,576-token shared context window, 65,536 default synchronous output limit, and up to 1,048,576 output tokens with background long decode. The Gemini API pricing page, captured 2026-10-14, lists Standard at $2/M input, $0.10/M cached input, and $10/M output through November 30, 2026, then $4/$0.20/$20 from December 1; Batch and Flex are half of Standard, and Priority is 1.8x. Caching is implicit only and Google says explicit caching is unsupported. A full Azure Retail Prices Foundry feed sweep and exact `gemini` meter search on 2026-10-14 returned no Google/Gemini token meter, so this is a Direct API lane with no Foundry rate inferred.