Update provider model pricing

Set ShareAI provider model input, output and cache pricing tiers using a scoped Provider API token and an exact model identifier.

View as Markdown

On this page

POST https://api.shareai.now/api/v1/console/provider/model-catalog/my/model/price

Update permitted pricing for a provider model.

Base URL: https://api.shareai.now
Authentication: HTTP Basic: provider UUID and provider access token
Required scope: provider_models:write

Use the catalog to identify the exact model and its provider-facing settings.

Pass full_model_identifier in the query string. The JSON body contains the pricing configuration. First read the current model and keep a copy of its prices so you can compare the saved result.

Example pricing body#

JSON

{
  "incentive_types": [
    "rewards",
    "exchange"
  ],
  "input": [
    {
      "min_tokens_inclusive": 0,
      "price_per_million": "0.95"
    }
  ],
  "output": [
    {
      "min_tokens_inclusive": 0,
      "price_per_million": "1.90"
    }
  ],
  "cache": [
    {
      "min_tokens_inclusive": 0,
      "price_per_million": "0.10"
    }
  ],
  "activation_mode": "immediate"
}

These are illustrative amounts, not recommended rates. Use exact decimal strings, and check the model’s pricing unit and eligible incentive settings in Console.

cURL

curl --fail-with-body --request POST \
  "https://api.shareai.now/api/v1/console/provider/model-catalog/my/model/price?full_model_identifier=URL_ENCODED_MODEL_IDENTIFIER" \
  -u "$PROVIDER_UUID:$PROVIDER_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  --data @pricing.json

Understand the fields#

Field	Meaning
`input / output / cache`	Separate tier lists for the corresponding token category.
`min_tokens_inclusive`	Token threshold at which a tier begins.
`price_per_million`	Exact amount per million tokens in the pricing contract.
`incentive_types`	Eligible incentive modes for this configuration.
`activation_mode`	When the supported configuration should take effect; the example uses immediate.

Verify the saved result#

Read the model again after the write and compare the effective pricing and activation state. A response acknowledging an update is not a reason to assume every active request changed price. Keep existing tasks and subsequent admissions separate.

Last updated September 15, 2026