Google Gemini
by Google
Access Gemini AI models for multimodal understanding, reasoning, and generation
How agents use Google Gemini
- ✓Agent uses Gemini as its reasoning engine for text generation, summarization, and question answering
- ✓Agent processes images, documents, or screenshots alongside text prompts in a multimodal step
- ✓Agent generates or reviews code as part of a development or automation workflow
- ✓Agent uses Gemini Flash for high-speed, low-latency inference in time-sensitive workflows
- ✓Agent leverages Gemini's large context window to analyze entire documents or codebases in one pass
Agent actions
Inputs: prompt, model, systemInstruction, maxOutputTokens, temperature
Returns: text, usage, finishReason
Inputs: messages, model, systemInstruction
Returns: text, usage, finishReason
Inputs: image, prompt, model
Returns: text, usage
Inputs: text, model, taskType
Returns: embedding, dimensions
Inputs: prompt, functions, model
Returns: functionCall, arguments
Inputs: prompt, model
Returns: result, code, output
Inputs: prompt, schema, model
Returns: json, usage
Example workflows
Multimodal assistant
Process text and images with Gemini
Code generation
Generate and execute code with Gemini
About Google Gemini
- Vendor
- Pricing Always review details with the vendor
- Free / Paid — Free tier available with rate limits. Gemini 1.5 Flash: 15 RPM free. Gemini 1.5 Pro: 2 RPM free. Higher usage requires paid tier.
- Authentication
- API key
- Rate limit Always review details with the vendor
- 60 requests / minute
- Compatible nodes
- AgentResourceOutput
Put Google Gemini behind your AI agent
Set up Google Gemini alongside your other tools in the planner — no code — so AI can act on it for your customers.
Open the planner