LLM
Gemini
Google Gemini chat, vision, and video models on your canvas — structured JSON output, streaming responses, and frame-level reasoning as drag-and-drop nodes.
What you can do
Bring Gemini 2.5 Flash, 2.5 Flash Lite, and the Gemini 3.x image and video models into your workflows. Chat, vision, image generation, and Veo-powered video, all available as drag-and-drop nodes.
- Chat completions with system prompts
- Structured JSON output
- Vision and image analysis
- Gemini Video node for frame-level reasoning
- Streaming responses supported
Used in these workflows
All use casesContent teams, marketing, indie creatorsContent → VoiceTurn a writing brief into a narrated audio asset in one flow. Script generation with Gemini, narration with ElevenLabs, returned to the caller via the published webhook.Support teams, customer success, opsAI Customer Support BotDraft support responses from webhook events. Classify tickets with Gemini or OpenAI and return a suggested reply your help desk can render or auto-send.Ops, finance, legal, back-officeDocument ProcessingUpload documents, extract structured JSON with vision-capable LLMs, return the result for your system of record.Product teams, AI engineers, researchersMulti-model comparisonRun the same prompt across Gemini and OpenAI in one flow. Compare cost, latency, and output quality on real prompts, not benchmark averages.E-commerce, media, content opsImage pipelinesResize, analyze, caption, and tag images. Deterministic resizing plus a vision-capable Gemini node, returned together for your catalog.
Start building your first flow today.
Free to try. No credit card required. Publish production workflows in under 10 minutes.