Developer Management

Deploy Cloud Model

Provision a cloud GPU instance, start a model serving runtime on it, and register the deployment as a cloud inference endpoint.

post/v1/developer/{developer_id}/deploy-cloud-model

Request

  • Base URL: https://api.oneinfer.ai (the document declares no server; this is the origin it was published from)
  • URL: https://api.oneinfer.ai/v1/developer/{developer_id}/deploy-cloud-model
  • Auth: HTTP bearer

Path parameters

developer_idstring required

Request body

provider_namestring required
instance_namestring required
gpu_idstring nullable
gpu_numinteger
disk_sizeinteger
image_urlstring required
envsobject[] nullable

Environment variables required by the image

commandstring nullable

Container command or arguments

start_commandstring nullable

Catalog alias for the container command

kindstring nullable

Novita instance kind

image_sourcestring nullable

Image source: catalog or registry

registry_urlstring nullable

Container registry URL or host

registry_imagestring nullable

Full private registry image path

registry_usernamestring nullable

Container registry username

registry_passwordstring nullable

Container registry password or access token

registry_auth_namestring nullable

Friendly name for saved registry credentials

registry_credential_idstring nullable

Saved OneInfer registry credential ID

registry_auth_idstring nullable

Provider registry auth ID

container_registry_auth_idstring nullable

Runpod container registry auth ID

regionstring required
startup_scriptstring nullable

Startup script to run on instance launch

additional_volume_sizenumber nullable

Additional volume size in GB (Verda: volume, Novita AI: network storage)

additional_volume_regionstring nullable

Volume Region

volume_namestring nullable

Name of existing volume to attach

volume_regionstring nullable

Region hint for an existing external volume

compute_typestring nullable

Compute type (e.g. CPU, GPU)

plan_idstring nullable

Provider plan ID for CPU instances

cpu_flavor_idstring nullable

CPU flavor ID for CPU instances

is_spotboolean nullable

Provision as a spot instance (Verda only). Uses spot pricing for credit checks.

endpoint_namestring required

Inference endpoint display name

model_source'catalog' | 'huggingface' | 'custom'
model_idstring nullable

Hugging Face model id or URL

model_template_idstring nullable

OneInfer model-template id

custom_model_s3_pathstring nullable

Uploaded custom-model S3 path

serving_librarystring nullable

Serving runtime used to expose an OpenAI-compatible endpoint

hf_access_tokenstring nullable

Optional Hugging Face access token for gated/private models

scaling_type'queue' | 'request_count'
max_concurrencyinteger
min_workersinteger
max_workersinteger
endpoint_urlstring nullable

Optional known endpoint URL. If omitted, backend will infer it when possible.

top_pnumber nullable
temperaturenumber nullable
max_tokensinteger nullable

Response

Successful Response

unknown required

Changes

No recorded changes to this operation across all 2 revisions of this API.