Wildbots Gemini Chat Model icon

Wildbots Gemini Chat Model

Google Gemini Chat Model via Cloudflare proxy — works in any region

Wildbots Gemini Chat Model

AI-generated

Summary

Generate chat completions using a selected Google Gemini language model via a configurable Cloudflare proxy.

Inputs

  • modelName (required) — Select the Google Gemini model to generate the chat completion, dynamically loaded from the models endpoint excluding embedding models.
  • options.maxOutputTokens — Specify the maximum number of tokens to generate in the completion (default 1024).
  • options.safetySettings — Configure zero or more safety categories (e.g., Hate Speech, Harassment) with thresholds to apply to the completion output.
  • options.temperature — Adjust randomness in token sampling; lower values yield less random, more deterministic completions (default 0.7).
  • options.topK — Limit the maximum number of tokens considered during sampling (default 40).
  • options.topP — Set the cumulative probability threshold for token sampling (default 0.9).

Output shape

a single AI language model instance configured to generate chat completions

The output is a language model object configured with the specified parameters and safety settings, ready for downstream nodes to invoke chat completion methods.

Links

Discussion