POST
/
v1
/
fine_tuning
/
create
curl --request POST \
  --url https://api.vlm.run/v1/fine_tuning/create \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "id": "<string>",
  "created_at": "2023-11-07T05:31:56Z",
  "callback_url": "<string>",
  "model": "<string>",
  "training_file": "<string>",
  "validation_file": "<string>",
  "num_epochs": 8,
  "batch_size": 123,
  "learning_rate": 0.0002,
  "suffix": "<string>",
  "wandb_base_url": "<string>",
  "wandb_api_key": "<string>",
  "wandb_project_name": "<string>"
}'
{
  "usage": {
    "elements_processed": 123,
    "element_type": "image",
    "credits_used": 123
  },
  "id": "<string>",
  "created_at": "2023-11-07T05:31:56Z",
  "completed_at": "2023-11-07T05:31:56Z",
  "status": "pending",
  "message": "<string>",
  "model": "<string>",
  "suffix": "<string>",
  "wandb_base_url": "<string>",
  "wandb_project_name": "<string>",
  "wandb_url": "<string>"
}

Authorizations

​
Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

Request to the VLM API for fine-tuning a model.

​
model
string
required

The model to fine-tune.

​
training_file
string
required

The file-id or URL of the training dataset.

​
id
string

Unique identifier of the request.

​
created_at
string

Date and time when the request was created (in UTC timezone)

​
callback_url
string | null

The URL to call when the request is completed.

Minimum length: 1
​
validation_file
string | null

The file-id or URL of the validation dataset.

​
num_epochs
integer
default:
8

The number of epochs to fine-tune the model.

​
batch_size
default:
8

The batch size to use for fine-tuning.

​
learning_rate
number
default:
0.0002

The learning rate to use for fine-tuning.

​
suffix
string | null

The suffix to use for the fine-tuned model.

​
wandb_base_url
string | null
default:
https://api.wandb.ai

The base URL to use for Weights & Biases.

​
wandb_api_key
string | null

The API key to use for Weights & Biases.

​
wandb_project_name
string | null

The project to use for the fine-tuning job.

Response

200
application/json
Successful Response

Response to the VLM API for fine-tuning a model.

​
message
string
required

The message from the fine-tuning job.

​
model
string
required

The model identifier to fine-tune (has the format ft:vlm-1:<my-org>:<custom_suffix>:id).

​
usage
object

The usage metrics for the request.

​
id
string

The unique identifier for the fine-tuning job.

​
created_at
string

The date and time when the job was created (in UTC timezone).

​
completed_at
string | null

The date and time when the job was completed (in UTC timezone).

​
status
enum<string>
default:
pending

The status of the fine-tuning job.

Available options:
enqueued,
pending,
running,
completed,
failed,
paused
​
suffix
string | null

The suffix to use for the fine-tuned model.

​
wandb_base_url
string | null
default:
https://api.wandb.ai

The base URL to use for Weights & Biases.

​
wandb_project_name
string | null

The project to use for the fine-tuning job.

​
wandb_url
string | null

The URL to the Weights & Biases project.

List FilesFinetuned Inference