Integrations
Official GenAI SDKs for Python, JavaScript, Go, Java and C++, plus the Gemini API with function calling and Search and Maps grounding for connecting external tools and data.
Learn moreBrowser workspace for prototyping with Gemini and shipping via API
Own Google AI Studio? Verify ownership to get a dofollow link

Google AI Studio is a web workspace for testing Gemini, Nano Banana, Veo, Lyria and other Google models, building apps from prompts, and moving to production through the Gemini API, SDKs and agent tools.
Google AI Studio is a browser-based workspace where Google's generative models can be tried, prompted and turned into working software. The welcome page groups the models by job: Gemini 4 for reasoning, Nano Banana for image generation, Veo and Omni for video creation and editing, Gemini TTS for controllable speech, Gemini Live for real-time voice and video conversations, and Lyria for music. Each one opens in a prompt or dedicated workspace, so a model can be tested before any code is written. The second half of the product is aimed at developers who want to move from experiment to production. A keys page handles creating Gemini API keys and monitoring usage. Official SDKs cover Python, JavaScript, Go, Java and C++. An Interactions API supports stateful, multi-turn agents with managed tool execution and history, and a tools layer offers function calling plus built-in support such as Search and Maps grounding. The same site hosts the documentation for all of these. AI Studio also works as an app builder. A text prompt, such as a language learning partner, a background remover for product photos or a garden design helper, starts a generated app. A gallery of projects can be remixed instead, including a multiplayer snake game, a window view generator based on live weather and a city management game driven by Gemini tasks. Case studies from re:cap, Calm Studios and AddGlow show how outside teams have built on the API. The audience is mainly developers, prototypers and technical product teams who already lean toward Google's model stack. Compared with standalone chat assistants, it exposes more of the model surface, with real keys, SDKs and agent tooling. Compared with neutral multi-model platforms, it only offers Google's own models. The homepage does not state pricing, usage limits or plan tiers, so cost and quota details need to be checked on Google's own documentation before committing a project to it.
Official GenAI SDKs for Python, JavaScript, Go, Java and C++, plus the Gemini API with function calling and Search and Maps grounding for connecting external tools and data.
Learn moreModel playgrounds for Gemini, Nano Banana, Veo, Omni, Gemini TTS, Gemini Live and Lyria. Prompt-to-app builder and remixable gallery. Gemini API key management, official SDKs, Interactions API for stateful agents, and tools including function calling with Search and Maps grounding.
Learn more