**This endpoint allows a CSV upload containing up to one million contacts or 5GB of data, whichever is smaller.** Imports take place asynchronously: the endpoint returns a URL (`upload_uri`) and HTTP headers (`upload_headers`) which can subsequently be used to `PUT` a file of contacts to be imported into our system. Uploaded CSV files may also be [gzip-compressed](https://en.wikipedia.org/wiki/Gzip). In either case, you must include the field `file_type` with the value `csv` in your request body. The `field_mappings` paramter is a respective list of field definition IDs to map the uploaded CSV columns to. It allows you to use CSVs where one or more columns are skipped (`null`) or remapped to the contact field. For example, if `field_mappings` is set to `[null, "w1", "_rf1"]`, this means skip column 0, map column 1 to the custom field with the ID `w1`, and map column 2 to the reserved field with the ID `_rf1`. See the "Get All Field Definitions" endpoint to fetch your custom and reserved field IDs to use with `field_mappings`. Once you recieve the response body you can then initiate a **second** API call where you use the supplied URL and HTTP header to upload your file. For example: `curl --upload-file "file/path.csv" "URL_GIVEN" -H 'HEADER_GIVEN'` If you'd like to monitor the status of your import job, use the `job_id` and the "Import Contacts Status" endpoint. Twilio SendGrid recommends exporting your contacts regularly as a backup to avoid issues or lost data.