Google Gemini
Gemini chat, embeddings, Imagen images, Veo video, speech, files and batch.
Route Gemini API models through DeepintShield with your own Gemini credentials. Vertex AI is a separate provider.
What works
- Chat Completions and Responses API, both with streaming
- Embeddings, speech (TTS) and transcription (STT)
- Image generation and edit with Gemini and Imagen; video generation with Veo
- Files, batch, named cached content and list models
- Multimodal input (text, images, video, PDF; audio through file references), code execution, tools and reasoning
Limits to know: No image variation, and no streaming for image generation or edit. Video delete, list and remix are not supported. TTS output is WAV only. Add google to the prompt-cache providers to use Gemini prompt caching.
How to connect
Add your Gemini API key in Model Hub → AI Providers, create a Virtual Key, then call gemini/<model> through https://app.deepintshield.com/v1 with any OpenAI-compatible client, or use the Google GenAI SDK integration.
Read the Gemini provider docs. Requests get the same Virtual Key budgets, guardrails and logs as every other provider on the AI gateway.