GET
/v1/schemas/reactor/video/reactor%2Fvisko-orbis-dynamic
Accept:
Request body
0 properties · 0 required · 0 $defsThis schema isn’t a plain object with properties — showing the raw JSON instead.
application/json
{
"provider": "reactor",
"activity": "video",
"endpointId": "reactor/visko-orbis-dynamic",
"kind": "input",
"contentHash": "df511cf755e0ca887a838fa42d3f5fc5b6beb62fce90b7e7521c23399c811351",
"specRevision": null,
"provenance": {
"sourceUrl": "https://reactor.inc/models/visko-orbis-dynamic/api#reactor/visko-orbis-dynamic",
"sourceHash": "099eafb92baa0a721d3fb3d58707fceb0e19f83c2ee20d63d5c233ad4fb23cf3",
"fetchedAt": 1789535490,
"extractorVersion": "1",
"derivation": "upstream-spec",
"verifiedAt": null
},
"createdAt": 1789535490,
"supersededAt": null,
"schema": {
"oneOf": [
{
"type": "object",
"required": [
"command",
"data"
],
"additionalProperties": false,
"properties": {
"command": {
"type": "string",
"const": "pause"
},
"data": {
"type": "object",
"properties": {}
}
}
},
{
"type": "object",
"required": [
"command",
"data"
],
"additionalProperties": false,
"properties": {
"command": {
"type": "string",
"const": "reset"
},
"data": {
"type": "object",
"properties": {}
}
}
},
{
"type": "object",
"required": [
"command",
"data"
],
"additionalProperties": false,
"properties": {
"command": {
"type": "string",
"const": "start"
},
"data": {
"type": "object",
"properties": {}
}
}
},
{
"type": "object",
"required": [
"command",
"data"
],
"additionalProperties": false,
"properties": {
"command": {
"type": "string",
"const": "resume"
},
"data": {
"type": "object",
"properties": {}
}
}
},
{
"type": "object",
"required": [
"command",
"data"
],
"additionalProperties": false,
"properties": {
"command": {
"type": "string",
"const": "set_seed"
},
"data": {
"type": "object",
"properties": {
"seed": {
"type": "integer",
"default": 42,
"minimum": 0,
"description": "Seed for the noise the first chunk is sampled from. Must be a non-negative integer; the model never draws its own seed, so the same seed with the same prompts reproduces the same video. Read once when `start` fires — later changes take effect only after `reset` followed by a new `start`.",
"x-reactor-moderate": false
}
}
}
}
},
{
"type": "object",
"required": [
"command",
"data"
],
"additionalProperties": false,
"properties": {
"command": {
"type": "string",
"const": "set_image"
},
"data": {
"type": "object",
"properties": {
"image": {
"$ref": "#/$defs/ReactorUploadReference",
"default": null,
"description": "Starting frame, uploaded via the Reactor presigned-URL protocol. PNG or JPEG. Any size — it is resized to the model's 640x368 without cropping, so a frame of a different aspect ratio is squashed rather than trimmed.",
"x-reactor-moderate": false
}
}
}
}
},
{
"type": "object",
"required": [
"command",
"data"
],
"additionalProperties": false,
"properties": {
"command": {
"type": "string",
"const": "set_prompt"
},
"data": {
"type": "object",
"properties": {
"prompt": {
"type": "string",
"default": "",
"description": "Natural-language description of the scene to generate. Replaces the previously active prompt. Applied on the next chunk when generating; otherwise takes effect when `start` fires.",
"x-reactor-moderate": true
},
"passthrough": {
"type": "boolean",
"default": false,
"description": "Send this prompt to the model exactly as written. This deployment prepares `set_prompt` text server-side first; true skips that, and the text sent is what conditions generation, byte for byte — for prompts already authored the way the model wants them. Decided per call, so prepared and verbatim prompts mix freely in one session — a prompt sent this way still carries its characters and setting into the next one. Inert on a deployment that prepares nothing, where every prompt is already verbatim.",
"x-reactor-moderate": false
}
}
}
}
},
{
"type": "object",
"required": [
"command",
"data"
],
"additionalProperties": false,
"properties": {
"command": {
"type": "string",
"const": "set_resolution"
},
"data": {
"type": "object",
"properties": {
"resolution": {
"type": "string",
"default": "",
"description": "One of the deployment's offered resolutions, exactly as listed in the `state` snapshot's `available_resolutions` — `native` is the model's generated geometry, the named tiers are upscaler-delivered (1080p = 1920x1080, 2k = 2560x1440, 4k = 3840x2160).",
"x-reactor-moderate": false
}
}
}
}
},
{
"type": "object",
"required": [
"command",
"data"
],
"additionalProperties": false,
"properties": {
"command": {
"type": "string",
"const": "set_audio_prompt"
},
"data": {
"type": "object",
"properties": {
"prompt": {
"type": "string",
"default": "",
"description": "What the scene should SOUND like — instruments, voices, materials, ambience. Not a description of what is on screen: sending a scene description here makes the audio worse than leaving it empty. Keep it to about one sentence; roughly the first 128 tokens are used and the rest is dropped without warning. Example: \"Acoustic guitar strums a rhythmic melody, with soft finger noise on the strings and quiet room ambience.\" Empty clears it, and the audio is then generated from the picture alone.",
"x-reactor-moderate": true
}
}
}
}
},
{
"type": "object",
"required": [
"command",
"data"
],
"additionalProperties": false,
"properties": {
"command": {
"type": "string",
"const": "set_audio_enabled"
},
"data": {
"type": "object",
"properties": {
"audio_enabled": {
"type": "boolean",
"default": true,
"description": "True to generate sound on `main_audio` (the default), false to skip the audio model and deliver silence from the next `start` on.",
"x-reactor-moderate": false
}
}
}
}
}
],
"$defs": {
"ReactorUploadReference": {
"type": "object",
"format": "reactor-upload-reference",
"required": [
"upload_id",
"name",
"mime_type",
"size"
],
"properties": {
"name": {
"type": "string"
},
"size": {
"type": "integer"
},
"mime_type": {
"type": "string"
},
"upload_id": {
"type": "string",
"format": "uuid"
}
},
"description": "Reference to a file uploaded via the Reactor upload protocol."
}
}
}
}Validate a payload
before you spend tokensshell
$ curl -X POST https://modelschemas.com/v1/validate \
-d '{"provider":"reactor","endpointId":"reactor/visko-orbis-dynamic","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}