`GET /v1/models`

Read the selected owned resource or current workflow state without starting a new generation.

Request example#

curl 'https://platform.tongai.io/v1/models' \  -H "Authorization: Bearer $TONG_API_KEY"

Configuration

curl --version

Authentication#

Use a server-side API key for your account, or OAuth for each customer’s account. Never send keys in prompts or tool arguments.

catalog:read

Parameters#

modelstringOptional

Choose a public Tong model ID from current discovery.

operationstringOptional

Select the workflow supported by the chosen model.

Request body#

This endpoint has no request body.

Response schema#

Response

JSON

{  "workflows": []}

`HTTP 200`

application/json

text_modelsobject[]Optional

Show nested fields

idstringRequired

namestringRequired

context_tokensintegerRequired

Maximum: 9007199254740991.

max_output_tokensintegerRequired

Maximum: 9007199254740991.

unitstringRequired

Allowed values"THOUSAND_TOKENS"

input_creditsstringRequired

A decimal string with six decimal places, for example 1.250000.

output_creditsstringRequired

A decimal string with six decimal places, for example 1.250000.

cached_input_creditsstringRequired

A decimal string with six decimal places, for example 1.250000.

cache_write_creditsstringRequired

A decimal string with six decimal places, for example 1.250000.

pricingstringRequired

Allowed values"display_maximum"

requires_exact_quotebooleanRequired

Allowed valuestrue

workflowsobject[]Required

Maximum items: 10000.

Show nested fields

operationstringRequired

modelstringRequired

namestringRequired

model_namestringRequired

resolutionstring | nullRequired

unitstringRequired

creditsstring | nullRequired

A decimal string with six decimal places, for example 1.250000.

base_creditsstring | nullRequired

A decimal string with six decimal places, for example 1.250000.

requires_exact_quotebooleanOptional

Default: false.

pricingstringOptional

Allowed values"configuration_quote""display_rate"

Default: "display_rate".

discount_percentnumberRequired

Minimum: 0.

Maximum: 100.

Errors and limits#

HTTP	Error	Recovery
`400`	Invalid request`REQUEST_INVALID`	Correct the fields identified by the error before trying again.
`401`	Invalid credential`CREDENTIAL_INVALID`	Check the credential, its expiry and whether it has been revoked.
`402`	Insufficient credits`INSUFFICIENT_CREDITS`	Add credits to the account that owns the credential, then continue the saved work.
`403`	Permission denied`SCOPE_REQUIRED`	Use a credential with the required scopes and access to the selected resource.
`404`	Resource unavailable`RESOURCE_UNAVAILABLE`	Check the resource ID and that it belongs to the authorized account.
`409`	Intent conflict`IDEMPOTENCY_CONFLICT`	Reuse an intent only with its original request. Use a new UUID for a different action.
`429`	Rate limited`RATE_LIMITED`	Wait for Retry-After before retrying the same intended request.
`503`	Temporarily unavailable`DEPENDENCY_UNAVAILABLE`	Wait for Retry-After and retry with the same intent. Check the saved operation after a timeout.

AuthenticationAPI keys charge the account that owns the key. OAuth lets each connected user use their own credits.Errors and rate limitsUnderstand error responses, retry safely and recover from failed generations.Requests and logsAn accepted request is not a finished generation. Track the saved job until its output is ready.