Azure OpenAI by YOOV icon

Azure OpenAI by YOOV

Interact with Azure OpenAI API, powered by YOOV
Join our community

Actions7

Overview

The Azure OpenAI by YOOV node, specifically for the Completions → Create operation, allows users to generate text completions using Azure-hosted OpenAI models (such as GPT-3.5 Turbo). This node is ideal for scenarios where you need AI-generated text based on a prompt, such as drafting emails, generating code snippets, answering questions, or creating conversational agents.

Practical examples:

  • Auto-generating customer support replies.
  • Summarizing long documents.
  • Creating creative writing prompts or story continuations.
  • Generating code or technical documentation from comments.

Properties

Name Meaning
API Version The version of the Azure OpenAI API to use. Options: 2023-07-01-preview, 2023-06-01-preview, 2023-05-15, 2023-03-15-preview, 2022-12-01, 2022-03-01-preview.
Model in Current Deployment The deployed model to use for completion. Options: gpt-35-turbo, gpt-35-turbo-16k, text-embedding-ada-002.
Prompt The input text or prompt for the model to complete. Can be a string, list of strings, or list of token lists. If not specified, defaults to <|endoftext
Max Tokens The maximum number of tokens to generate. Minimum: 0.
Temperature Sampling temperature; higher values produce more creative results. Range: 0–2.
Top P Nucleus sampling parameter; considers tokens with top_p probability mass. Range: 0–1.
Logit Bias JSON object mapping token IDs to bias values (-100 to 100) to influence likelihood of specific tokens appearing. Example: {"50256": -100} prevents <|endoftext
User ID of the end-user for tracking and rate-limiting.
N Number of completions to generate per prompt. Range: 1–128.
Stream Whether to enable streaming responses (server-sent events).
Logprobs Include log probabilities for the most likely tokens. Range: 0–100.
Suffix Text to append after the generated completion.
Echo Whether to echo back the prompt in addition to the completion.
Stop Sequence that indicates the end of the current document. Default: <|endoftext
Completion Config Additional configuration for completion (string).
Cache Level Server-side caching control: 0 = no cache, 1 = prompt prefix enabled, 2 = full cache.
Presence Penalty Penalizes new tokens based on their frequency so far. Range: -2 to 2.
Frequency Penalty Penalizes new tokens if they appear in the text so far. Range: -2 to 2.
Best Of Number of generations to create server side, displaying only the best. Max: 128. No streaming if >1.
Model Name of the model to use (string).

Output

The output is a JSON object containing the response from the Azure OpenAI Completions API. Typical fields include:

{
  "id": "unique-request-id",
  "object": "text_completion",
  "created": 1234567890,
  "model": "gpt-35-turbo",
  "choices": [
    {
      "text": "Generated completion text.",
      "index": 0,
      "logprobs": null,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 10,
    "completion_tokens": 20,
    "total_tokens": 30
  }
}
  • id: Unique identifier for the request.
  • object: Type of object returned (e.g., "text_completion").
  • created: Timestamp of creation.
  • model: Model used for generation.
  • choices: Array of generated completions, each with:
    • text: The generated text.
    • index: Position in the choices array.
    • logprobs: Log probabilities (if requested).
    • finish_reason: Why the completion finished ("stop", "length", etc.).
  • usage: Token usage statistics.

Note: The actual structure may vary depending on the API version and options selected.

Dependencies

  • External Service: Requires access to Azure OpenAI API.
  • Credentials: You must configure the yoovAzureOpenAIApi credential in n8n, which should include your Azure OpenAI endpoint and API key.
  • n8n Configuration: Ensure the node has network access to Azure services.

Troubleshooting

Common Issues:

  • Invalid Credentials: If the API key or endpoint is incorrect, you'll receive authentication errors. Double-check your credentials in n8n.
  • Model Not Deployed: If the selected model is not available in your Azure deployment, the API will return an error. Verify your Azure OpenAI resource and deployment settings.
  • Prompt Too Large: Exceeding the maximum allowed prompt size will result in an error. Reduce the prompt length or number of prompts.
  • Parameter Errors: Supplying invalid types (e.g., non-numeric value for "Max Tokens") can cause validation errors.

Error Messages:

  • "Resource not found": Check if the model name and deployment are correct.
  • "Invalid API version": Ensure the selected API version matches those supported by your Azure OpenAI resource.
  • "Authentication failed": Recheck your API key and endpoint.
  • "Input exceeds maximum context length": Shorten your prompt or reduce the number of tokens requested.

Links and References

Discussion