A restaurant menu dataset you can try.
Download 164 Baja Fresh records in CSV or JSON. Read the ingredient text and check whether the fields fit your project before you subscribe.
Free for internal evaluation and noncommercial research. Read the sample terms.
What the files contain
This sample comes from the database snapshot collected on September 17, 2026 at 15:00 UTC. Each row keeps its own source collection date in scraped_at. The sample is a fixed selection from one restaurant; it does not establish current coverage of the restaurant’s whole menu.
| Field | How to read it |
|---|---|
| item_id, restaurant_id | Identifiers for joining records. |
| ingredients_text | Collected wording, including component lists and possible alternatives. Oils and additives appear in this text. |
| nutrition, nutrition_method | Nutrient values and how the pipeline obtained them. Some records use sums of menu components. Null means unknown. |
| allergens | Collected labels, raw disclosure and source status. Review the raw text; labels can need cleanup and do not establish allergy safety. |
| provenance.source_urls | Source links when stored. This sample has empty source-link lists. |
| record_observed_at, scraped_at | Database observation time and source collection time. |
CSV cells contain JSON-encoded values. After parsing the CSV, use JSON.parse on each cell to restore strings, lists, objects and nulls. The schema version is pruf.restaurant_database.v1.
Try a search and check the limits
Our ingredient search example runs on these files and explains how missing source links and alternative ingredients affect the result.
The sample terms allow selected factual excerpts with PRūF attribution and the sample date. A paid subscription allows commercial applications within the customer terms. Bulk redistribution requires a separate written agreement.
Get API access and monthly releases
Data costs $199 per month. Your subscription includes the covered restaurant dataset, downloads and the menu explorer.
View coverage and pricing