Actions18
- ChatGPT Actions
- Claude Actions
- Gemini Actions
- Keyword Data Actions
- LLM Mentions Actions
- Perplexity Actions
Gemini → Get Live Gemini LLM Responses
AI-generatedSummary
Sends a prompt to the specified Google Gemini AI model and returns the structured response.
Inputs
- user_prompt (required) — The question or task to send to the AI model. Limited to 500 characters.
- model_name (required) — The name of the Gemini model (e.g., gemini-1.5-pro). If a base name is used, the latest version is applied automatically.
- max_output_tokens — Maximum number of tokens for the AI response. Optional.
- temperature — Randomness of the response. Range: 0 to 2. Optional.
- top_p — Diversity of the response. Range: 0 to 1. Optional.
- web_search — Enable web search to enhance the response. Defaults to false. Optional.
- system_message — Instructions defining the AI's role, tone, or behavior. Optional.
- message_chain (required) — Array of previous messages to include in the conversation history. Each object requires a 'role' ('user' or 'ai') and a 'message'.
Output shape
The raw JSON response from the DataForSEO API, containing the structured output of the Gemini model.
The API returns a response envelope (status_code, tasks array). The actual content is typically nested within tasks[0].result[0].
Examples
Example 1: Simple single-turn question
Set user_prompt to 'What is 2+2?', model_name to 'gemini-1.5-flash', and leave other parameters at defaults.
Example 2: Conversation with system instructions
Set system_message to 'You are a helpful coding assistant.' and message_chain to include previous turns (e.g., user: 'Write a loop' -> ai: 'Here is the loop...').