Google adds Veo and Gemini image models to OpenAI compatibility layer

Google

Mar 28, 2026 · Updated Apr 25, 2026

Google expanded its OpenAI compatibility layer to include Veo video generation and Gemini image models. Developers can now switch between providers by changing only three lines of code in their existing Python or JavaScript pipelines. This removes the technical friction of testing Google's multimodal capabilities against industry standards.

Google updated its OpenAI compatibility layer to support Veo video generation and Gemini image models. Developers using the OpenAI Python or JavaScript SDKs can now access veo-3.1-generate-preview, gemini-2.5-flash-image, and gemini-3-pro-image-preview. The integration requires minimal changes, specifically updating the api_key, base_url, and model parameters.

This update targets the high switching costs of proprietary SDKs. By adopting the OpenAI API specification for multimodal models, Google allows teams to benchmark performance against competitors with zero code rewrites. It also introduces a standardized way to handle long-running video generation tasks through a Sora-compatible /v1/videos endpoint.

You can now trigger generation using familiar workflows, passing parameters like aspect_ratio or safety_settings through the extra_body field. Video generation returns an operation ID for polling until the status is completed. These features are in beta and available via the Gemini API using a standard API key.

View the full update on ai.google.dev

Google AI Developers

@googleaidevsMar 26

Veo and Gemini image models such as Nano Banana are now live in the OpenAI compatibility layer! Test our models in your existing Python/JS pipelines with zero SDK rewrite. Just update exactly 3 lines of code (𝚊𝚙𝚒_𝚔𝚎𝚢, 𝚋𝚊𝚜𝚎_𝚞𝚛𝚕, 𝚖𝚘𝚍𝚎𝚕). https://t.co/C0eIZq4jqg

View on X

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

See all AI news & updates from Google →

Keep reading

Google releases Veo 3.1 Lite for high volume video at half cost

Google launched Veo 3.1 Lite, a new video generation model available through the Gemini API and Google AI Studio that supports text-to-video and image-to-video workflows. By offering the same speed as the Fast model at less than half the price, this update makes high-volume video applications economically viable for developers.

Google DeepMindMay 20

Google DeepMind Launches Gemini Omni to Reimage and Edit Video Content

Google DeepMind introduced Gemini Omni Flash, a multimodal model that allows users to transform existing video scenes using natural language prompts. By combining generative media systems with Gemini's reasoning, the model can instantly swap environments or add objects while maintaining the original video's action.

VercelFeb 19

Vercel AI Gateway Adds Video Generation Across Four Providers

Vercel AI Gateway added video generation with models from xAI, Kling, Alibaba Wan, and Google Veo, all callable through generateVideo() in AI SDK 6. Switching providers is a model string change - no separate integrations, auth flows, or dashboards needed.

Google AI StudioMar 18

Gemini API Adds Per-Project Monthly Spend Caps in AI Studio

Google launched Project Spend Caps for the Gemini API, letting developers set a monthly dollar limit per project in AI Studio. Caps have up to a 10-minute activation delay, and Google revamped usage tiers to auto-upgrade developers as usage scales.