API reference · Schema
FineTuneBaseModel
objectstringrequiredAlways `fine_tune_base_model`: the type of this object
One of
fine_tune_base_modelidstringrequiredUnique ID (`mdl_…`)
ID with prefix mdl_
modelstringrequiredModel ID (`publisher/name`)
display_namestringrequiredName to show people
licensestring | nullrequiredLicence of the model's weights; null when not recorded
methodsarray of stringrequiredTraining methods the model supports
adaptersarray of stringrequiredAdapter types the model supports
max_seq_lengthintegerrequiredLongest example in tokens; longer ones are truncated
hardwarearray of stringrequiredSKUs it trains on (first = default)
priceTrainingPrice | any JSON | nullrequiredThe training price per token in the price book in force; null when fine-tuning this model is not priced (creating a job is refused)
ShowHide 5 fields of
priceunit_pricestringrequiredInteger nano-USD as a decimal string
pattern ^-?[0-9]+$
unit_quantitystringrequiredTokens the unit price is for
per_million_tokensstringrequiredInteger nano-USD as a decimal string
pattern ^-?[0-9]+$
display_unitstring | nullrequiredLabel of the unit for display; null when none is set
price_book_versionintegerrequiredVersion of the price book the price comes from
pricing_placeholderbooleanrequiredTrue while the model's prices are development placeholders
{
"object": "fine_tune_base_model",
"id": "mdl_01jc5zq4m8k3v6x9t2w7r0b1e4",
"model": "string",
"display_name": "string",
"license": "string",
"methods": [
"supervised"
],
"adapters": [
"lora"
],
"max_seq_length": 1,
"hardware": [
"string"
],
"price": {
"unit_price": "1500000000",
"unit_quantity": "string",
"per_million_tokens": "1500000000",
"display_unit": "string",
"price_book_version": 1
},
"pricing_placeholder": true
}Used by
- Models that can be fine-tuned, how, on which hardware and at what price
GET /v1/fine-tunes/base-models