General FAQ

Frequently Asked Questions

What AI models are used?

Docfire AI Engine uses Google Gemini 2.5 Flash as its primary model. The system is designed to support other models (Claude, GPT) as fallback options in the future.

How long does processing take?

Processing time depends on the number of products and the template used. On average:

• Translate: 1-5 seconds per batch of 10 products
• PDF Extraction: 5-15 seconds per document
• SEO Description: 2-8 seconds per batch of 10 products

Is there a limit on the number of products per request?

There is no hard limit, but it is recommended to send batches of no more than 50 products per request for optimal results.

Is data retained?

Session data and results are stored in the PostgreSQL database. Uploaded PDF files are temporarily saved in local storage.

How can I monitor processing status?

Use the GET /api/sessions/{id}/status endpoint to check the status. Implement a polling mechanism with an interval of 2-5 seconds.