AI / Google Antigravity Gemini Fundamentals Interview Questions
What is the Gemini API and what does it give developers access to?
The Gemini API is Google's developer interface to its Gemini family of multimodal AI models. It allows developers to integrate state-of-the-art language, vision, audio, image generation, and video capabilities into applications via simple HTTP calls or official SDKs.
Access is through Google AI Studio (developer portal, free tier available) or Google Cloud Vertex AI (enterprise, higher limits, compliance features). All requests require an API key passed as the x-goog-api-key header.
| Category | Capabilities |
|---|---|
| Language models | Text generation, reasoning, coding, multilingual understanding |
| Multimodal understanding | Process images, video, audio, PDFs together with text |
| Image generation | Nano Banana (Gemini image) models for generation and editing |
| Video generation | Veo 3.1 models for cinematic video with native audio |
| Agents | Managed agents including Antigravity, Deep Research |
| Grounding | Search Grounding for real-time web data |
| Speech | Text-to-speech and Live API for audio conversations |
# Install the Python SDK pip install google-genai # Quickstart from google import genai client = genai.Client() # reads GEMINI_API_KEY from env # Interactions API (recommended for new projects as of June 2026) interaction = client.interactions.create( model="gemini-3.5-flash", input="Explain recursion in one sentence.", ) print(interaction.output_text)
More Related questions...