Register a model

Registers a new model owned by your account. Callable by an account_admin or developer when model serving is enabled for your account. The CLI equivalent is reactor model register.

model.name is <org>/<name>, where <org> is your account’s slug, or a bare <name>, which is placed under your account’s slug. Your account must have a slug. Names are unique within an account, and a deleted model keeps its name reserved.

New models are private. A customer cannot make a model public; contact Reactor for that.

status is an integer on this request (1 active, 2 deprecated, 3 disabled; default active), while every response returns it as a lowercase string. gpu_type takes a string name from GET /gpu_types.

POST

models

curl --request POST \
  --url https://api.reactor.inc/models \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": {
    "name": "your-org/your-model"
  },
  "description": "Real-time video model",
  "is_gpu": true,
  "gpu_type": "NVIDIA_H100",
  "gpu_count": 1
}
'

{
  "id": "7b3f1bc2-a4e5-4d78-b9c1-0e6c2f9a1d34",
  "account_id": "8c8b7e12-a4c1-4627-b0a0-5d2e9f3a7b61",
  "account": "Your Org",
  "name": "your-org/your-model",
  "description": "Real-time video model",
  "status": "active",
  "is_public": false,
  "is_gpu": true,
  "gpu_type": "NVIDIA_H100",
  "gpu_count": 1,
  "created_at": "2025-05-11T21:46:40Z",
  "updated_at": "2025-05-11T21:46:40Z"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

Authorizations

Authorization

string

header

required

A Reactor API key (rk_...), sent as Authorization: Bearer rk_.... Keep it on a server.

Body

application/json

model

object

required

Show child attributes

description

string

Human-facing description.

status

enum<integer>

default:1

Initial status as an integer: 1 active, 2 deprecated, 3 disabled. Responses return the lowercase string.

Available options:

1,

2,

3

is_public

boolean

default:false

Must be false or omitted; a customer cannot create a public model.

is_gpu

boolean

default:false

Whether the model needs a GPU. With true, gpu_type is required.

gpu_type

string

An accelerator name from GET /gpu_types. Integer values are refused.

Example:

"NVIDIA_H100"

gpu_count

integer

GPUs per instance. Some types allow only 1.

Required range: x >= 0

extra_args

string[]

Extra command-line arguments passed to the model container.

cpu_request

string

CPU request of the model container, such as 4. Defaults apply when omitted.

cpu_limit

string

CPU limit; must be at least cpu_request.

memory_request

string

Memory request, such as 32Gi.

memory_limit

string

Memory limit; must be at least memory_request.

workers

object

Makes the model multi-session: each instance runs one model server, which holds the model and the GPUs, plus count workers, each serving one session. The model's own gpu_*, cpu_*, and memory_* fields apply to the model server. Omit workers for one session per instance. The CLI sends this from model.resources.workers in reactor.yaml.

Show child attributes

Response

The model was created. The body is the model detail, without project_id.

The full view of a model, for its owner.

string<uuid>

required

name

string

required

Example:

"your-org/your-model"

status

enum<string>

required

The model's status, always lowercase in responses.

Available options:

active,

deprecated,

disabled

is_public

boolean

required

is_gpu

boolean

required

created_at

string<date-time>

required

When it was created, as an RFC 3339 timestamp.

Example:

"2026-01-15T10:30:00Z"

updated_at

string<date-time>

required

When it was last updated, as an RFC 3339 timestamp.

Example:

"2026-01-15T10:30:00Z"

account_id

string<uuid>

The owning account.

account

string

The owning account's name.

description

string

gpu_type

string

Example:

"NVIDIA_H100"

gpu_count

integer

extra_args

string[]

cpu_request

string

Example:

"4"

cpu_limit

string

memory_request

string

Example:

"32Gi"

memory_limit

string

workers

object

Show child attributes

pod_total

object

read-only

For a model with workers: the CPU and memory of one whole instance, with defaults filled in. This is what your account's per-model quotas are checked against.

Show child attributes

project_id

string<uuid>

The model's project. Only on GET /models/{id}.

⌘I

curl --request POST \
  --url https://api.reactor.inc/models \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": {
    "name": "your-org/your-model"
  },
  "description": "Real-time video model",
  "is_gpu": true,
  "gpu_type": "NVIDIA_H100",
  "gpu_count": 1
}
'

{
  "id": "7b3f1bc2-a4e5-4d78-b9c1-0e6c2f9a1d34",
  "account_id": "8c8b7e12-a4c1-4627-b0a0-5d2e9f3a7b61",
  "account": "Your Org",
  "name": "your-org/your-model",
  "description": "Real-time video model",
  "status": "active",
  "is_public": false,
  "is_gpu": true,
  "gpu_type": "NVIDIA_H100",
  "gpu_count": 1,
  "created_at": "2025-05-11T21:46:40Z",
  "updated_at": "2025-05-11T21:46:40Z"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

Get started

SDK Reference

CLI Reference

Platform API Reference

Endpoints

Register a model

Authorizations

Body

Response