Start training from a version

POST

/v1/projects/{projectId}/versions/{versionNumber}/trainings

curl -X POST 'https://api.embodiflow.io/v1/projects/{projectId}/versions/{versionNumber}/trainings' \
  -H 'Authorization: Bearer $EMBODIFLOW_PAT' \
  -H 'Content-Type: application/json' \
  -d '{
    "policy": "act",
    "computeFlavor": "string",
    "trainingSteps": 0,
    "preset": "quick",
    "advanced": {
      "hyperparameters": {},
      "baseModelId": "00000000-0000-0000-0000-000000000000",
      "seed": 0,
      "imageRevisionId": "00000000-0000-0000-0000-000000000000",
      "coTrainVersion": "00000000-0000-0000-0000-000000000000"
    }
  }'

Authorization

Send a personal access token with every request.

Authorization: Bearer PAT

Authentication guide

Path parameters

projectIdstringrequired
versionNumbernumberrequired

Header parameters

Idempotency-Keystring
Optional idempotency key, at most 255 characters (UUIDv4 recommended). A repeated key returns the first response byte-for-byte, including a 5xx. The same key with a different body, or a second in-flight request, returns 409 IDEMPOTENCY_CONFLICT. Keys are unique per tenant and retained for 24 hours.

Request body

Content-Typeapplication/json

policystringrequired
Allowed values
act
diffusion
smolvla
pi0_lora
pi05_lora
groot
wall_oss
computeFlavorstringrequired
trainingStepsnumber
presetstring
Allowed values
quick
standard
long
advancedobject
- hyperparametersobject
- baseModelIdstring (uuid)
- seednumber
- imageRevisionIdstring (uuid)
- coTrainVersionstring (uuid)

Responses

Success

dataobjectrequired
- modelIdstringrequired
- trainingJobIdstringrequired
- statestringrequired

Response example

{
  "data": {
    "modelId": "string",
    "trainingJobId": "string",
    "state": "string"
  }
}

Servershttps://api.embodiflow.io/v1https://api.embodiflow.dev/v1