/api/orgs/{org_id}/provider/applyApply to become a provider
Files the org's application for platform review. When called with an API key and models is non-empty, every declared wire id must first complete a streamed completion against base_url (a house_key applicant must send smoke_api_key, which authenticates the probe only and is never stored); a failure answers 400 and nothing is filed. An empty models list applies now and adds lanes after approval. 409 when the org already operates a provider, already has a pending application, or the slug is taken. Org admin session or the org's own key.
Authentication: Bearer API key in the Authorization header
Path parameters
org_idstringrequiredYour organization id (uuid).
GET /api/whoamireturns it.
Request body
slugstringrequireddisplay_namestringrequiredbase_urlstringrequiredPublic https OpenAI-compatible base URL, no query or fragment.
auth_modeenumrequiredhouse_keyif your endpoint needs a key,nonefor an open endpoint.logo_urlstring | nullPublic https URL of your logo.
modelsarray<object>First lanes; each must declare
data_policy. Empty applies without lanes.models
wire_idstringrequiredThe model id your endpoint accepts in
model.bindingobjectrequiredBind to a catalog model (
existing) or propose a new one (new).One of
object
kindstringrequiredslugstringrequiredAn existing catalog model slug.
object
kindstringrequiredmodelobjectrequiredmodel
slugstringrequireddisplay_namestringrequireddescriptionstring | nullcontext_windowinteger | nullmax_output_tokensinteger | nullinput_modalitiesarray<enum>huggingface_urlstring | nullExactly https://huggingface.co//.
release_urlstring | nullhttps URL with a real dotted host.
pricesobjectrequiredPer-million-token rates in integer nano-USD (1 USD = 1,000,000,000 nano-USD). Each rate is at most 1,000,000,000,000 ($1,000 per million tokens).
prices
input_nano_usd_per_millionintegerrequiredoutput_nano_usd_per_millionintegerrequiredcached_input_nano_usd_per_millioninteger | nullCached input rate; null means no cached rate.
reasoning_nano_usd_per_millioninteger | nulleffective_atstring | nullOptional timestamp with a timezone offset.
capabilitiesobjectDeclared capability booleans. Keys must come from the gateway's capability vocabulary (for example
supports_streaming,supports_parallel_tool_calls,supports_structured_text,supports_image_input,reports_cached_input_tokens,reports_reasoning_tokens); unknown keys are refused. A lane needssupports_streaming: trueto be listed.context_windowinteger | nullregionstring | nulldeprecationobject | nulldeprecation
eol_atstringrequiredreplacement_slugstring | null
data_policyobject | nullPer-model data handling. Required on every model in an application.
One of
object
retains_promptsbooleanrequiredretention_daysinteger | nullDays prompts are kept.
trains_on_databooleanrequiredzero_data_retentionbooleanrequiredpolicy_urlstring | nullPublic https URL of the policy.
dispatchobject | nullOne of
object
scale_up_secondsinteger | nullWarm-up time before saturation counts against the lane.
overflow_policyenumWhat happens once saturated.
max_concurrencyinteger | nullDeclared cap on in-flight requests for this model.
rpminteger | nullRequests per minute dispatch limit for the new lane; omit for the default.
tpminteger | nullTokens per minute dispatch limit for the new lane; omit for the default.
contactobjectrequiredcontact
emailstringrequiredWhere the platform reaches you when a lane misbehaves.
escalationstring | nullFree-form escalation instructions (pager, shared channel, phone).
status_pagestring | nullPublic https status page.
notestring | nullsmoke_api_keystring | nullEndpoint key for the server-side smoke an API-key applicant with declared models must pass. Never stored.
Responses
201The filed application.
Response
idstringrequiredslugstringrequireddisplay_namestringrequiredbase_urlstringrequiredauth_modeenumrequiredlogo_urlstring | nullrequiredmodelsarray<object>requiredmodels
wire_idstringrequiredThe model id your endpoint accepts in
model.bindingobjectrequiredBind to a catalog model (
existing) or propose a new one (new).One of
object
kindstringrequiredslugstringrequiredAn existing catalog model slug.
object
kindstringrequiredmodelobjectrequiredmodel
slugstringrequireddisplay_namestringrequireddescriptionstring | nullcontext_windowinteger | nullmax_output_tokensinteger | nullinput_modalitiesarray<enum>huggingface_urlstring | nullExactly https://huggingface.co//.
release_urlstring | nullhttps URL with a real dotted host.
pricesobjectrequiredPer-million-token rates in integer nano-USD (1 USD = 1,000,000,000 nano-USD). Each rate is at most 1,000,000,000,000 ($1,000 per million tokens).
prices
input_nano_usd_per_millionintegerrequiredoutput_nano_usd_per_millionintegerrequiredcached_input_nano_usd_per_millioninteger | nullCached input rate; null means no cached rate.
reasoning_nano_usd_per_millioninteger | nulleffective_atstring | nullOptional timestamp with a timezone offset.
capabilitiesobjectDeclared capability booleans. Keys must come from the gateway's capability vocabulary (for example
supports_streaming,supports_parallel_tool_calls,supports_structured_text,supports_image_input,reports_cached_input_tokens,reports_reasoning_tokens); unknown keys are refused. A lane needssupports_streaming: trueto be listed.context_windowinteger | nullregionstring | nulldeprecationobject | nulldeprecation
eol_atstringrequiredreplacement_slugstring | null
data_policyobject | nullPer-model data handling. Required on every model in an application.
One of
object
retains_promptsbooleanrequiredretention_daysinteger | nullDays prompts are kept.
trains_on_databooleanrequiredzero_data_retentionbooleanrequiredpolicy_urlstring | nullPublic https URL of the policy.
dispatchobject | nullOne of
object
scale_up_secondsinteger | nullWarm-up time before saturation counts against the lane.
overflow_policyenumWhat happens once saturated.
max_concurrencyinteger | nullDeclared cap on in-flight requests for this model.
rpminteger | nullRequests per minute dispatch limit for the new lane; omit for the default.
tpminteger | nullTokens per minute dispatch limit for the new lane; omit for the default.
data_policyobjectrequiredSelf-declared prompt-data handling.
retention_daysis required whenretains_promptsis true and must be omitted otherwise;zero_data_retentionrequiresretains_promptsandtrains_on_databoth false.data_policy
retains_promptsbooleanrequiredretention_daysinteger | nullDays prompts are kept.
trains_on_databooleanrequiredzero_data_retentionbooleanrequiredpolicy_urlstring | nullPublic https URL of the policy.
contactobjectrequiredcontact
emailstringrequiredWhere the platform reaches you when a lane misbehaves.
escalationstring | nullFree-form escalation instructions (pager, shared channel, phone).
status_pagestring | nullPublic https status page.
notestring | nullrequiredstatusenumrequiredreview_notestring | nullrequiredcreated_atstringrequiredreviewed_atstring | nullrequired
400The server-side smoke failed (
code: smoke_failed) or ahouse_keykey applicant declared models withoutsmoke_api_key.Response
errorstringrequiredHuman-readable message.
codestringStable machine reason, when the route defines one (for example
pro_required).actionstringStable remediation action, when the route defines one.
detailsobjectCustomer-safe string facts the caller needs to act (for example a suggested identifier).
401Missing, unknown or revoked API key.
Response
errorstringrequiredHuman-readable message.
codestringStable machine reason, when the route defines one (for example
pro_required).actionstringStable remediation action, when the route defines one.
detailsobjectCustomer-safe string facts the caller needs to act (for example a suggested identifier).
403The key is valid but may not perform this action (for example an inference key on a provisioning-key route, or a feature outside your plan).
Response
errorstringrequiredHuman-readable message.
codestringStable machine reason, when the route defines one (for example
pro_required).actionstringStable remediation action, when the route defines one.
detailsobjectCustomer-safe string facts the caller needs to act (for example a suggested identifier).
404The resource does not exist or belongs to another organization.
Response
errorstringrequiredHuman-readable message.
codestringStable machine reason, when the route defines one (for example
pro_required).actionstringStable remediation action, when the route defines one.
detailsobjectCustomer-safe string facts the caller needs to act (for example a suggested identifier).
409The org already operates a provider, already has a pending application, or the slug is registered.
Response
errorstringrequiredHuman-readable message.
codestringStable machine reason, when the route defines one (for example
pro_required).actionstringStable remediation action, when the route defines one.
detailsobjectCustomer-safe string facts the caller needs to act (for example a suggested identifier).
422The body or query failed validation.
Response
detailarray<object>requireddetail
typestringlocarray<string | integer>msgstring
Request
curl -X POST https://api.experientiallabs.ai/api/orgs/3f6c2a1e-8b7d-4c2a-9e51-6a0d4b2f7c10/provider/apply \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"slug": "acme-inference",
"display_name": "Acme Inference",
"base_url": "https://api.acme-inference.example/v1",
"auth_mode": "house_key",
"logo_url": null,
"models": [
{
"wire_id": "meta-llama/llama-4.1-70b-instruct",
"binding": {
"kind": "existing",
"slug": "llama-4.1-70b-instruct"
},
"prices": {
"input_nano_usd_per_million": 350000000,
"output_nano_usd_per_million": 900000000
},
"capabilities": {
"supports_streaming": true
},
"context_window": 131072,
"data_policy": {
"retains_prompts": false,
"trains_on_data": false,
"zero_data_retention": true
}
}
],
"contact": {
"email": "oncall@acme-inference.example",
"escalation": "PagerDuty: acme-llm",
"status_page": "https://status.acme-inference.example"
},
"note": "H200 capacity in us-east, 99.9% uptime target.",
"smoke_api_key": "sk-acme-probe-7d2c9e4b1a6f"
}'Response
{
"id": "4e7b1c93-2a6d-4f58-9b0e-d3c5a8f2e716",
"slug": "acme-inference",
"display_name": "Acme Inference",
"base_url": "https://api.acme-inference.example/v1",
"auth_mode": "house_key",
"logo_url": null,
"models": [
{
"wire_id": "meta-llama/llama-4.1-70b-instruct",
"binding": {
"kind": "existing",
"slug": "llama-4.1-70b-instruct"
},
"prices": {
"input_nano_usd_per_million": 350000000,
"output_nano_usd_per_million": 900000000,
"cached_input_nano_usd_per_million": null,
"reasoning_nano_usd_per_million": null,
"effective_at": null
},
"capabilities": {
"supports_streaming": true
},
"context_window": 131072,
"region": null,
"deprecation": null,
"data_policy": {
"retains_prompts": false,
"retention_days": null,
"trains_on_data": false,
"zero_data_retention": true,
"policy_url": null
},
"dispatch": null,
"rpm": null,
"tpm": null
}
],
"data_policy": {
"retains_prompts": false,
"retention_days": null,
"trains_on_data": false,
"zero_data_retention": true,
"policy_url": null
},
"contact": {
"email": "oncall@acme-inference.example",
"escalation": "PagerDuty: acme-llm",
"status_page": "https://status.acme-inference.example"
},
"note": "H200 capacity in us-east, 99.9% uptime target.",
"status": "pending",
"review_note": null,
"created_at": "2026-11-19T10:30:02.551Z",
"reviewed_at": null
}https://api.experientiallabs.ai/api/orgs/3f6c2a1e-8b7d-4c2a-9e51-6a0d4b2f7c10/provider/apply