Zhipu AI (GLM)Install

Summary

Send a required user prompt to a selected Zhipu chat model, optionally guided by a system message. Vision models can also receive an image URL; the operation returns the API response and, when present, the first choice’s message content and usage data.

Inputs

• Model (required) — Select a chat model; the default is GLM-4.7-Flash.
• Prompt (required) — The user message sent to the model.
• System Message — Optional message that guides the model’s behavior.
• Image URL — For the listed vision models, provide an image URL to send with the prompt. The source includes it only when the selected model is vision-capable; otherwise it sends the prompt as text.
• Additional Fields — Optional generation settings: sampling (Do Sample), maximum tokens, request ID, comma-separated stop sequences, streaming, temperature, top P, and end-user ID. Stop sequences are split on commas and trimmed.

Output shape

One output item per input item, each containing a response object with the full API response; it may also include text and usage fields.

The text field is taken from the first choice’s message content when present. The usage field is included when returned by the API. If Continue On Fail is enabled, failed items instead contain an error field.

Links

• n8n community nodes installation guide