/api/modelsCreate a custom model
Creates an organization-owned model row, its deployments (always customer_managed, billed to your own provider or endpoint) and its default chain. Idempotent: an exact replay answers 200 with the existing rows; the same slug with different attributes answers 409, and a public model's slug answers 409 model_slug_reserved (add a deployment to the public model instead). A deployment with a base_url needs a verified payment method and counts toward 10 new endpoints per hour.
認証: AuthorizationヘッダーにBearer形式のAPIキー
リクエストボディ
org_idstring | nullOwning organization. Implied by a key; a session caller must name one.
slugstring必須Letter first, lowercase alphanumerics with single
.,_or-separators. Becomes the model name requests send.display_namestring必須descriptionstring | nullrelease_datestring | nullcontext_windowinteger | nullmax_output_tokensinteger | nullinput_modalitiesarray<enum>output_modalitiesarray<enum>supported_paramsobjectParameters the model supports, for example
{"temperature": true, "tools": true}.categorystring | nulltagsarray<string>providersarray<object>必須providers
providerenum必須provider_model_idstring必須base_urlstring | nullRequired for
localandmodal; forbidden for fixed-origin providers. An explicit http(s) endpoint with optional port and path, no userinfo, query or fragment.regionstring | nullapi_versionstring | nullprovider_connection_idstring | nullPin to one of your API-key BYOK connections of the same provider; null resolves by provider at request time.
input_nano_usd_per_millioninteger | nullInteger nano-USD per million tokens; null = unknown.
cached_input_nano_usd_per_millioninteger | nullcache_write_input_nano_usd_per_millioninteger | nullcache_write_1h_input_nano_usd_per_millioninteger | nulllong_context_cache_write_input_nano_usd_per_millioninteger | nulllong_context_cache_write_1h_input_nano_usd_per_millioninteger | nulloutput_nano_usd_per_millioninteger | nullreasoning_nano_usd_per_millioninteger | nullpricing_sourcestring | nullWhere the rates come from (for example
self-reported).provider-free-tieris reserved and refused.capabilitiesobjectWire capabilities. Seeded from the model's
supported_params; explicit keys win.uptime_30dnumber | nullthroughput_tpsnumber | nulllatency_p50_msnumber | nullstats_sourceenumendpoint_api_keystring | nulllocalonly. Sent to your endpoint asAuthorization: Bearer <key>. Stored in a vault, never returned.
レスポンス
201Created. A replay of an identical request answers 200 with the same body.
レスポンス
modelobjectOne catalog model row.
model
idstringslugstringThe name a request sends as
model.display_namestringdescriptionstring | nullrelease_datestring | nullISO date.
context_windowinteger | nullAdvertised window in tokens. For most makers it is a total that includes the maximum output.
max_output_tokensinteger | nullmax_input_tokensinteger | nullThe prompt budget a caller can actually send (the window minus the maximum output when the window is a total). Pin client compaction to this, not
context_window.input_modalitiesarray<enum>output_modalitiesarray<enum>supported_paramsobjectLogical parameters the model supports, for example
{"tools": true}.categorystring | nulltagsarray<string>iconstring | nullMaker family key used for the icon.
makerstring | nullowning_org_idstring | nullYour organization id for a custom model; null for a public model.
preferred_rankinteger | nullCurated rank (lower first); null when unranked.
statusstringactiveorhidden.created_atstringupdated_atstring
providersarray<object>providers
idstringDeployment id, or
byok:{connection_id}for a lane served by your own provider connection. This is the token a waterfall write uses.model_idstringproviderstringProvider id as shown to you; platform lanes run on our own capacity show
experiential_cloud.provider_model_idstringThe provider's wire id.
base_urlstring | nullEndpoint of a customer-run (
local,modal) deployment.regionstring | nullapi_versionstring | nullowning_org_idstring | nullYour organization for a private lane; null for a public lane.
provider_connection_idstring | nullThe pinned BYOK connection, if any.
connection_aliasstring | nullOn BYOK lanes only: the connection's account alias, which tells two same-provider accounts apart.
route_idstring | nullOpaque per-organization handle for selecting exactly this lane on a request. Null without an organization context or when the lane cannot serve your traffic.
provider_statusobjectThe provider's own public status report for this lane. Advisory only, never routing input.
provider_status
stateenumofficial_urlstring | nullchecked_atstring | nullscopestringlimitationstringincidentsarray<object>advisoryboolean
provider_logo_urlstringRegistry provider logo, when one exists. Omitted otherwise.
billing_sourceenumhost_managedbills your credits;customer_managedbills your own provider or endpoint.input_nano_usd_per_millioninteger | nullcached_input_nano_usd_per_millioninteger | nullcache_write_input_nano_usd_per_millioninteger | nullcache_write_1h_input_nano_usd_per_millioninteger | nulllong_context_cache_write_input_nano_usd_per_millioninteger | nulllong_context_cache_write_1h_input_nano_usd_per_millioninteger | nullcache_write_12h_input_nano_usd_per_millioninteger | nullNull on a lane supporting 12-hour cache means 2.5 times the input rate.
output_nano_usd_per_millioninteger | nullreasoning_nano_usd_per_millioninteger | nullimage_output_nano_usd_per_millioninteger | nullGenerated-image tokens rate; null bills them as output.
long_context_threshold_tokensinteger | nullOnce reported input tokens reach this, the
long_context_*rates price the whole request. Null means flat pricing.long_context_input_nano_usd_per_millioninteger | nulllong_context_cached_input_nano_usd_per_millioninteger | nulllong_context_output_nano_usd_per_millioninteger | nulllong_context_reasoning_nano_usd_per_millioninteger | nullunit_pricesobjectA per-unit lane's only meter.
unit_prices
kindenumratesobjectInteger nano-USD per unit by variant (
""is the flat rate).
off_peak_pricingobjectOff-peak price card with its weekly UTC peak windows (
prices,peak_windows,off_peak_dates). The headline rates are the peak price. Omitted when absent.pricing_sourcestring | nullpricing_effective_atstring | nullcapabilitiesobjectWire capabilities the gateway enforces (for example
supports_tools,supported_reasoning_efforts).uptime_30dnumber | nullPercent.
uptime_recentnumber | nullCompleted share (percent) of the route's most recent requests; null under 50 requests.
uptime_recent_requestsinteger | nullthroughput_tpsnumber | nullObserved decode tokens per second.
latency_p50_msnumber | nullObserved median time to first token.
request_latency_p50_msnumber | nullObserved median end-to-end latency of calls with no first token.
stats_sourceenumstatusstringroutableboolean | nullWhether the gateway's active plan routes this platform lane. Null when not applicable.
requires_paymentbooleanPaid-only lane (a completed top-up opens it).
requires_purchasebooleanPurchase-required lane (a real purchase, not the $1 verification, opens it).
zdr_onlybooleanServes only requests routed under zero data retention.
hidden_fallbackbooleanavailable_to_orgboolean | nullWhether your organization may use this closed lane. Null without an organization context or for an open lane.
zero_data_retentionboolean | nullDetail route only. Null means no posture coverage, which a require-ZDR policy treats as non-compliant.
zero_data_retention_on_requestboolean | nullZDR only when dispatched under the gateway's per-request ZDR constraint.
processing_region_setsarray<array<string>>plan_discountsobject | nullいずれか
object
freenumberpronumbermaxnumberultranumber
service_tier_ratesobject | nullPer-tier rate cards (
flex,priority) in nano-USD per million tokens when the lane prices processing tiers.ladder_rankinteger | nullDisplay-only position among the model's public platform lanes.
ladder_lane_countinteger | nullladder_scorenumber | nullladder_factorsobject | nullendpoint_credential_last4string | nullLast four characters of a stored endpoint API key.
created_atstringupdated_atstring
default_waterfallarray<object>default_waterfall
idstringpositioninteger0 is tried first.
model_provider_idstring | nullprovider_connection_idstring | nulltarget_kindenumreferenced_model_idstring | nullreferenced_modelobject | nullいずれか
object
idstringslugstringdisplay_namestringstatusstring
providerstring | nullprovider_model_idstring | nullbase_urlstring | nullstatusstringrungstring | nullThe token to send in a PUT or PATCH: the deployment id,
byok:{connection_id}ormodel:{model_id}.fundingenumhealthobject | nullいずれか
object
statestringFleet verdict, for example healthy, degraded, recovering, throttled, circuit_open, disabled. A BYOK rung reports its connection status.
reasonstring | nulllifecycleenum
retentionenumZero data retention across the platform lanes: all, at least one, or none.
huggingface_urlstring | nullrelease_urlstring | nullbenchmarksarray<object>benchmarks
benchmarkstringdisplay_namestringunitstringhigher_is_betterbooleanscorenumbersourcestringsource_urlstring | nullretrieved_atstring
promotionsarray<object>promotions
labelstringlistedbooleanbadge_styleenumdisplay_onlyboolean
promotional_listedbooleanads_on_all_requestsbooleanSponsored ads may appear on every platform-funded request to this model.
401Missing, unknown or revoked API key.
レスポンス
errorstring必須Human-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 organization is under review (
org_under_review), or abase_urldeployment needs a verified payment method (endpoint_requires_verified_payment).レスポンス
errorstring必須Human-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).
404org_idnames an organization you cannot act in, orprovider_connection_idis not one of your connections.レスポンス
errorstring必須Human-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 slug exists with different attributes, or is a public model's slug (
model_slug_reserved).レスポンス
errorstring必須Human-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).
422Body validation failed, or a rule the schema enforces (for example
pricing_source_reserved, a connection of the wrong provider, orplan_connection_not_pinnable).429The organization created 10 custom endpoints in the last hour (
endpoint_creation_rate_limited).Retry-Aftergives the seconds to wait.レスポンス
errorstring必須Human-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).
リクエスト
curl -X POST https://api.experientiallabs.ai/api/models \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"slug": "my-local-model",
"display_name": "My Local Model",
"supported_params": {
"temperature": true,
"tools": true
},
"context_window": 131072,
"providers": [
{
"provider": "local",
"provider_model_id": "my-model",
"base_url": "https://models.acme.dev:8000/v1",
"endpoint_api_key": "sk-acme-endpoint-4f9a2c7e1b",
"input_nano_usd_per_million": 0,
"output_nano_usd_per_million": 0,
"pricing_source": "self-reported"
}
]
}'レスポンス
{
"model": {
"id": "0b8f3d2a-5c61-4e7f-9a12-7d4e6b1c9f30",
"slug": "my-local-model",
"display_name": "My Local Model",
"description": null,
"release_date": null,
"context_window": 131072,
"max_output_tokens": null,
"max_input_tokens": 131072,
"input_modalities": [
"text"
],
"output_modalities": [
"text"
],
"supported_params": {
"temperature": true,
"tools": true
},
"category": null,
"tags": [],
"icon": null,
"maker": null,
"owning_org_id": "3f6c2a1e-8b7d-4c2a-9e51-6a0d4b2f7c10",
"preferred_rank": null,
"status": "active",
"created_at": "2026-11-19T15:02:11.104512+00:00",
"updated_at": "2026-11-19T15:02:11.104512+00:00"
},
"providers": [
{
"id": "5d2e9a71-3b4c-4f80-8e16-2a9c7b0d4e53",
"model_id": "0b8f3d2a-5c61-4e7f-9a12-7d4e6b1c9f30",
"provider": "local",
"provider_model_id": "my-model",
"base_url": "https://models.acme.dev:8000/v1",
"region": null,
"api_version": null,
"owning_org_id": "3f6c2a1e-8b7d-4c2a-9e51-6a0d4b2f7c10",
"provider_connection_id": null,
"billing_source": "customer_managed",
"input_nano_usd_per_million": 0,
"cached_input_nano_usd_per_million": null,
"cache_write_input_nano_usd_per_million": null,
"cache_write_1h_input_nano_usd_per_million": null,
"long_context_cache_write_input_nano_usd_per_million": null,
"long_context_cache_write_1h_input_nano_usd_per_million": null,
"output_nano_usd_per_million": 0,
"reasoning_nano_usd_per_million": null,
"long_context_threshold_tokens": null,
"long_context_input_nano_usd_per_million": null,
"long_context_cached_input_nano_usd_per_million": null,
"long_context_output_nano_usd_per_million": null,
"long_context_reasoning_nano_usd_per_million": null,
"pricing_source": "self-reported",
"pricing_effective_at": "2026-11-19T15:02:11.104512+00:00",
"capabilities": {
"supports_temperature": true,
"supports_tools": true
},
"uptime_30d": null,
"throughput_tps": null,
"latency_p50_ms": null,
"stats_source": null,
"status": "active",
"endpoint_credential_last4": "7e1b",
"created_at": "2026-11-19T15:02:11.104512+00:00",
"updated_at": "2026-11-19T15:02:11.104512+00:00"
}
],
"default_waterfall": [
{
"id": "9c1a4e72-6d3b-4f85-b2e0-1f7a8c5d3e94",
"position": 0,
"model_provider_id": "5d2e9a71-3b4c-4f80-8e16-2a9c7b0d4e53",
"provider_connection_id": null,
"target_kind": "deployment",
"referenced_model_id": null,
"referenced_model": null,
"provider": "local",
"provider_model_id": "my-model",
"base_url": "https://models.acme.dev:8000/v1",
"status": "active",
"rung": "5d2e9a71-3b4c-4f80-8e16-2a9c7b0d4e53",
"funding": "own_endpoint",
"health": null
}
],
"retention": null,
"huggingface_url": null,
"release_url": null,
"benchmarks": [],
"promotions": [],
"promotional_listed": false,
"ads_on_all_requests": false
}https://api.experientiallabs.ai/api/models