API Release Notes

Gemini Decision (gemini-decision-1) and the Decisions API are available for fast, calibrated binary, categorical, and rubric decisions across text, images, audio, video, and PDFs. The API produces zero output tokens.

Gemini 4 Carbon (gemini-4-carbon) is generally available as the most capable reasoning model for sustained agentic coding, multimodal knowledge work, scientific reasoning, and cybersecurity defense. Introductory pricing is $2.00 per 1M input and $10.00 per 1M output tokens through November 30, 2026.

Requests to gemini-embedding-2-preview are deprecated and now route to gemini-embedding-2 at Gemini Embedding 2 rates. Update the model string; no other code changes or re-indexing are needed.

Gemini Omni 1.2 Flash is generally available with up to 15-second video generation, conversational editing, improved character consistency across up to six reference images, 30% lower video output pricing, and doubled paid-tier rate limits. Omni 1.1 Flash requests will auto-route to 1.2 Flash on December 3, 2026, and the API now accepts PPT, ODP, and ODS inputs plus audio, video, and fileData URIs in function responses.

The Gemini API now accepts PPT, ODP, and ODS files as input, with DOC, PPT, and ODP rendered as page images and ODS read from first-sheet data and formulas. Function responses can also return Office and OpenDocument files, audio, and video, with GenerateContent accepting fileData, displayName, and video mediaProcessing.

Prebuilt and designed voices can be modified with natural-language instructions through POST /v1beta/voices, producing new stored voice IDs for use with Gemini 3.8 Flash TTS and Flash-Lite TTS.

gemini-2.5-pro and gemini-2.5-flash are deprecated, with requests routing to gemini-3.8-flash and gemini-3.5-flash-lite respectively on December 1, 2026. Both come with breaking API changes: function calling and thought_signature validation, thinking_level replacing thinking_budget, and the removal of sampling parameters and prefilled model turns.

Computer Use is now generally available across the Interactions and GenerateContent APIs.

The Computer Use tool is generally available across the Interactions and GenerateContent APIs.

Released the gemini-api CLI in beta, a stateless, zero-dependency command-line client for AI coding agents, scripts, and CI pipelines. It uses the Interactions API to generate text and media, analyze files, and run Gemini managed agents.

The Gemini API CLI (gemini-api) is in beta, a stateless, zero-dependency command-line client for AI coding agents, scripts, and CI pipelines. It works against the Interactions API to generate text and media, analyze files, and run Gemini managed agents.

New projects using the Generate Content API on the paid tier now store requests by default with store=true, matching the Interactions API, while existing projects keep their current setting. Opt out via store=false per request or in the Logs and Datasets Settings panel.

New paid-tier projects using generateContent, streamGenerateContent, and batchGenerateContent now store requests by default (store=true), matching the Interactions API. Existing projects keep their current setting; opt out with store=false per request or by disabling Generate Content API logging in the Settings panel of the Logs and Datasets dashboard.

Gemini 4 Argon is generally available, positioned as the most capable model for complex coding, agentic, math, and science tasks. API changes, pricing, and migration guidance are covered in the "What's new in Gemini 4 Argon" documentation.

Gemini 3.7 Flash is deprecated and routed to gemini-3.8-flash, while gemini-3.5-flash is deprecated and routed to gemini-3.6-flash. The deep-research-pro-preview-12-2025 model shuts down on October 23, 2026.

Gemini Nano Banana 2.1 (gemini-nano-banana-2.1) is generally available with improved visual quality, prompt adherence, multi-turn character consistency, text rendering, and wide and panoramic aspect ratios up to 8:1 across 1K, 2K, and 4K resolutions. The gemini-3.1-flash-image model is deprecated with no shutdown date announced.

Gemini Nano Banana 2.1 is now generally available, and gemini-3.1-flash-image is deprecated.

Gemini 3.8 Flash TTS (gemini-3.8-flash-tts) and the faster, cheaper Gemini 3.8 Flash-Lite TTS (gemini-3.8-flash-lite-tts) are generally available alongside the Gemini API Voices endpoint at /v1beta/voices. Voice design, consent-verified voice replication, and an extended library of 150+ prebuilt and custom voices are included, with Flash-Lite TTS positioned to replace gemini-3.1-flash-tts-preview.

Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS are generally available, alongside the Gemini API Voices endpoint at /v1beta/voices. Flash-Lite TTS replaces gemini-3.1-flash-tts-preview for high-throughput production and real-time voice agent cascades, and voice design, voice replication with consent verification, and a 150+ voice library are now offered.

Access to Gemini 2.5 models is being limited to users who have actively used them in the past; the models are not deprecated and will continue to be served until further notice. New projects should use 3.5 Flash-Lite or 3.8 Flash.