Files

Upload

To upload a new dataset file:

tg files upload [FILENAME]

Here’s a sample output:

$ tg files upload ./example.jsonl
Uploading file example.jsonl: 100%|██████████████████████████████| 5.18M/5.18M [00:01<00:00, 4.20MB/s]
Success!
file-d931200a-6b7f-476b-9ae2-8fddd5112308

The printed file-… identifier is the assigned file-id for this file object. Pass --json to get the full response body instead.

Parameters

Flag	Description
`--purpose [string]`	The purpose of the file. Default: `fine-tune`. Must be one of: `fine-tune` `eval` `eval-sample` `eval-output` `eval-summary` `batch-generated` `batch-api`
`--no-check`	Skip local structural validation before uploading (UTF-8 and JSON object per non-empty line for JSONL). The server still validates during ingestion.

Check

You can precheck that a file is structurally valid before uploading.

tg files check [PATH]

Here’s a sample output:

$ tg files check ./local-file.jsonl
Validating file: 1 lines [00:00, 7476.48 lines/s]
OK Checks passed

Pass --json to get the full structured report instead of the pass/fail summary:

$ tg files check ./local-file.jsonl --json
{
  "is_check_passed": true,
  "message": "Checks passed",
  "found": true,
  "file_size": 620,
  "utf8": true,
  "line_type": true,
  "text_field": true,
  "key_value": true,
  "has_min_samples": true,
  "num_samples": 5,
  "load_json": true,
  "load_csv": null,
  "filetype": "jsonl"
}

For fine-tuning JSONL files, this command verifies UTF-8 encoding, that each non-empty line is parseable JSON, and that there is one JSON object per line, along with the minimum sample count and maximum file size. Schema-level validation (such as required fields and role ordering) is performed on the server when the file is used. The Files API surfaces any errors at that point.

List

To list previously uploaded files:

tg files list

Retrieve

To retrieve the metadata of a previously uploaded file:

tg files retrieve [FILE_ID]

Here’s a sample output:

$ tg files retrieve file-d931200a-6b7f-476b-9ae2-8fddd5112308
Retrieved file details
Id:      file-d931200a-6b7f-476b-9ae2-8fddd5112308 
Name:    info_provided_validation_tokenized.parquet
Size:    303.4 KB
Type:    parquet
Purpose: fine-tune
Created: 03/16/2026, 03:58 PM

Retrieve content

To download a previously uploaded file:

tg files retrieve-content <FILE_ID> \
  --output ./

tg files retrieve-content <FILE_ID> \
  --stdout

Here’s a sample output:

$ tg files retrieve-content file-d931200a-6b7f-476b-9ae2-8fddd5112308 --output ./
File saved to ./example.jsonl

You must pass either --output <DIRECTORY> to write the contents to disk under the file’s original filename, or --stdout to stream the contents to standard output.

Delete

To delete a previously uploaded file:

tg files delete [FILE_ID]

Here’s a sample output:

$ tg files delete file-d931200a-6b7f-476b-9ae2-8fddd5112308
√ File deleted

TOGETHER CLI

COMMANDS

Upload

Parameters

Check

List

Retrieve

Retrieve content

Delete

​Upload

​Parameters

​Check

​List

​Retrieve

​Retrieve content

​Delete

Upload

Parameters

Check

List

Retrieve

Retrieve content

Delete