Overview
This node integrates with Yandex's GPT language models to generate text completions based on user input. It is designed for scenarios where users want to leverage Yandex Foundation Models for natural language processing tasks such as content generation, summarization, question answering, or any other AI-driven text completion.
Practical examples include:
- Generating creative writing or marketing copy.
- Automating customer support responses.
- Creating summaries of long documents.
- Assisting in coding or data analysis by generating code snippets or explanations.
The node allows selecting a specific Yandex GPT model either by providing a direct URI or choosing from a list of available models.
Properties
| Name | Meaning |
|---|---|
| Model | The Yandex GPT model used to generate the completion. You can specify it by entering a model URI (e.g., gpt://<folder_id>/yandexgpt-lite/latest) or select from a searchable list of available models documented by Yandex. |
| Options | Additional parameters to customize the completion generation: - Temperature: Controls randomness in the output (0 excluded to 1). Lower values produce more analytical or deterministic results; higher values yield more creative outputs. - Max Tokens: Maximum number of tokens allowed in the generated completion. Must be greater than zero and may depend on the selected model. |
Output
The node outputs an object under the key Model containing the response from the Yandex GPT model. This includes the generated text completion result in JSON format.
If the node supports binary data output, it would represent related binary content from the model, but this implementation focuses on textual completions only.
Dependencies
- Requires an API key credential for authenticating with Yandex Cloud's Foundation Models service.
- The node depends on the external
@langchain/yandexpackage to interact with Yandex GPT models. - Proper configuration of the API key and folder ID within the credentials is necessary for successful requests.
Troubleshooting
- Invalid URI error: If the model URI does not match the expected pattern (
^(gpt|ds)://(.+)/(.+)), the node will reject it. Ensure the URI is correctly formatted or select a model from the provided list. - Authentication failures: Errors related to API key or folder ID usually indicate misconfigured credentials. Verify that the API key is valid and has access to the specified folder.
- Token limit exceeded: Setting
Max Tokenstoo high for the chosen model might cause errors. Adjust the value according to the model's documented limits. - Timeouts or network issues: These can occur if the Yandex API is unreachable. Check network connectivity and Yandex service status.