> ## Documentation Index > Fetch the complete documentation index at: https://apidoc.cometapi.com/llms.txt > Use this file to discover all available pages before exploring further. # Opprett en video med Grok > Opprett asynkrone Grok Imagine Video 1.5-oppgaver fra tekst, et kildebilde eller bilde- og forhåndsinnstilte stemmereferanser. Bruk dette endepunktet for å starte en Grok-videooppgave fra tekst, et kildebilde eller referanseinndata. Det returnerer umiddelbart en `request_id`, så behandle det som det første trinnet i en asynkron arbeidsflyt. Send alltid `model: grok-imagine-video-1.5` eksplisitt. `model` og `prompt` kreves for 1.5-forespørselsformene som dokumenteres her. ## Velg en inndatamodus | Modus | Forespørselsfelt | Regler | | ------------------- | ------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------- | | Tekst-til-video | `prompt` | Utelat `image`, `reference_images` og `reference_audios`. | | Bilde-til-video | `prompt`, `image.url` | `image.url` godtar en offentlig bilde-URL eller en bilde-data-URI. Ikke kombiner `image` med referanseinndata. | | Referanse-til-video | `prompt` og minst én av `reference_images` eller `reference_audios` | Send opptil 7 referansebilder og opptil 3 forhåndsinnstilte stemmer. Utdata fra referanse-til-video er begrenset til `720p`. | For en data URI bruker du `data:image/png;base64,`. Hvert `reference_images`-element bruker det samme `url`-feltet. Et referanselydelement bruker en forhåndsinnstilt `voice_id`, for eksempel `eve`. I Prompt kan tagger som `` og `` identifisere den tilsvarende referansen. ## Begynn med en liten forespørsel * Bruk `model: grok-imagine-video-1.5` * For den første forespørselen bør du holde `duration` på `1` og `resolution` på `480p` * Hold `prompt` eksplisitt slik at scenen og bevegelsen er tydelige * Legg til én inndatamodus om gangen før du kombinerer referansebilder og stemmer ## Varighet og oppløsning | Innstilling | Støttede verdier | Standard | Atferd ved grenseverdier | | -------------- | ------------------------------------------------- | -------------------------- | ---------------------------------------------------------------------------------------------------------------------- | | `duration` | heltall `1`-`15` | `8` | Bruk kortere klipp ved første tester; lengre klipp koster mer og tar lengre tid å fullføre. | | `resolution` | `480p`, `720p`, `1080p` | `480p` | Forespørsler med høyere oppløsning koster mer og tar lengre tid å fullføre. Referanse-til-video støtter opptil `720p`. | | `aspect_ratio` | `1:1`, `16:9`, `9:16`, `4:3`, `3:4`, `3:2`, `2:3` | `16:9` for tekst-til-video | For bilde-til-video bevarer utelatelse kildebildets sideforhold. | Genererte videoer inkluderer et lydspor. Angi `duration`, `resolution` og `aspect_ratio` eksplisitt når applikasjonen din avhenger av et bestemt utdataformat. ## Oppgaveflyt Send modellen, Prompt og eventuelle inndata, og lagre den returnerte `request_id`. Kall [Hent xAI-videoresultater](./get-video-generation-results) til `status` på toppnivå blir `done`, `failed` eller `expired`. Et første svar kan bare inneholde den gjentatte `request_id` mens oppgavemetadata klargjøres. Kopier den endelige `video.url` til din egen lagring hvis du trenger den etter det midlertidige leveringsvinduet. ## OpenAPI ````yaml api/openapi/video/xai/post-video-generation.openapi.json POST /grok/v1/videos/generations openapi: 3.1.0 info: title: Grok Video Generation API version: 1.0.0 description: >- Create an asynchronous Grok Imagine Video 1.5 job from text, a source image, or reference inputs. servers: - url: https://api.cometapi.com security: - bearerAuth: [] paths: /grok/v1/videos/generations: post: summary: Create a Grok video generation job description: >- Start a Grok Imagine Video 1.5 job. Save the returned `request_id`, then poll the result endpoint until the top-level status reaches a terminal state. operationId: video_generation requestBody: required: true content: application/json: schema: type: object additionalProperties: false required: - model - prompt allOf: - not: required: - image - reference_images - not: required: - image - reference_audios - if: anyOf: - required: - reference_images - required: - reference_audios then: properties: resolution: description: Reference-to-video resolution constraint. enum: - 480p - 720p properties: model: type: string enum: - grok-imagine-video-1.5 description: >- The Grok video model ID. Send this field explicitly so the request uses Grok Imagine Video 1.5. example: grok-imagine-video-1.5 prompt: type: string minLength: 1 description: >- A description of the scene, motion, and audio. Reference-to-video prompts can identify inputs with tags such as `` and ``. example: A paper boat glides across a quiet pond at sunrise. duration: type: integer minimum: 1 maximum: 15 default: 8 description: >- The output duration in seconds. Use an integer from 1 through 15. aspect_ratio: type: string description: >- The output aspect ratio. Text-to-video uses `16:9` when omitted. Image-to-video uses the source image's aspect ratio when omitted. enum: - '1:1' - '16:9' - '9:16' - '4:3' - '3:4' - '3:2' - '2:3' example: '16:9' resolution: type: string description: >- The output resolution. Reference-to-video requests support `480p` and `720p`; text-to-video and image-to-video also support `1080p`. enum: - 480p - 720p - 1080p default: 480p image: type: object additionalProperties: false required: - url properties: url: type: string description: >- A public image URL or an image data URI such as `data:image/png;base64,`. description: >- The source image for image-to-video. Do not combine this field with `reference_images` or `reference_audios`. reference_images: type: array minItems: 1 maxItems: 7 description: >- Visual references for reference-to-video. Each item accepts a public image URL or image data URI. Do not combine this field with `image`. items: type: object additionalProperties: false required: - url description: One visual reference. properties: url: type: string description: A public image URL or an image data URI. reference_audios: type: array minItems: 1 maxItems: 3 description: >- Preset voice references for reference-to-video. Do not combine this field with `image`. items: type: object additionalProperties: false required: - voice_id description: One preset voice reference. properties: voice_id: type: string minLength: 1 description: The preset voice ID, such as `eve`. examples: text-to-video: summary: Text to video value: model: grok-imagine-video-1.5 prompt: A paper boat glides across a quiet pond at sunrise. duration: 1 aspect_ratio: '16:9' resolution: 480p image-url: summary: Image URL value: model: grok-imagine-video-1.5 prompt: >- Animate the clouds with a slow camera move and natural motion. image: url: >- https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png duration: 1 aspect_ratio: '16:9' resolution: 720p image-data-uri: summary: Image data URI value: model: grok-imagine-video-1.5 prompt: Animate the clouds drifting across the sky. image: url: data:image/png;base64, duration: 1 aspect_ratio: '16:9' resolution: 480p reference-image: summary: Reference image value: model: grok-imagine-video-1.5 prompt: >- Use as the visual reference for a gentle cinematic pan. reference_images: - url: >- https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png duration: 1 aspect_ratio: '16:9' resolution: 720p reference-image-and-audio: summary: Reference image and preset voice value: model: grok-imagine-video-1.5 prompt: >- Use as the visual reference while says Welcome to the observatory. reference_images: - url: >- https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png reference_audios: - voice_id: eve duration: 1 aspect_ratio: '16:9' resolution: 720p responses: '200': description: Request accepted. content: application/json: schema: type: object additionalProperties: false required: - request_id properties: request_id: type: string description: >- The deferred request ID used to poll `GET /grok/v1/videos/{request_id}`. example: request_id: example: request_id: x-codeSamples: - lang: Shell label: Text to video source: |- curl https://api.cometapi.com/grok/v1/videos/generations \ -H "Authorization: Bearer $COMETAPI_KEY" \ -H "Content-Type: application/json" \ --data-binary '{ "model": "grok-imagine-video-1.5", "prompt": "A paper boat glides across a quiet pond at sunrise.", "duration": 1, "aspect_ratio": "16:9", "resolution": "480p" }' - lang: Python label: Text to video source: |- import os import requests payload = { "model": "grok-imagine-video-1.5", "prompt": "A paper boat glides across a quiet pond at sunrise.", "duration": 1, "aspect_ratio": "16:9", "resolution": "480p" } response = requests.post( "https://api.cometapi.com/grok/v1/videos/generations", headers={ "Authorization": "Bearer " + os.environ["COMETAPI_KEY"], "Content-Type": "application/json", }, json=payload, timeout=180, ) response.raise_for_status() print(response.json()["request_id"]) - lang: JavaScript label: Text to video source: >- const payload = { "model": "grok-imagine-video-1.5", "prompt": "A paper boat glides across a quiet pond at sunrise.", "duration": 1, "aspect_ratio": "16:9", "resolution": "480p" }; const response = await fetch( "https://api.cometapi.com/grok/v1/videos/generations", { method: "POST", headers: { Authorization: `Bearer ${process.env.COMETAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify(payload), }, ); if (!response.ok) throw new Error(`Request failed: ${response.status}`); console.log((await response.json()).request_id); - lang: Shell label: Image URL source: |- curl https://api.cometapi.com/grok/v1/videos/generations \ -H "Authorization: Bearer $COMETAPI_KEY" \ -H "Content-Type: application/json" \ --data-binary '{ "model": "grok-imagine-video-1.5", "prompt": "Animate the clouds with a slow camera move and natural motion.", "image": { "url": "https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png" }, "duration": 1, "aspect_ratio": "16:9", "resolution": "720p" }' - lang: Python label: Image URL source: |- import os import requests payload = { "model": "grok-imagine-video-1.5", "prompt": "Animate the clouds with a slow camera move and natural motion.", "image": { "url": "https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png" }, "duration": 1, "aspect_ratio": "16:9", "resolution": "720p" } response = requests.post( "https://api.cometapi.com/grok/v1/videos/generations", headers={ "Authorization": "Bearer " + os.environ["COMETAPI_KEY"], "Content-Type": "application/json", }, json=payload, timeout=180, ) response.raise_for_status() print(response.json()["request_id"]) - lang: JavaScript label: Image URL source: >- const payload = { "model": "grok-imagine-video-1.5", "prompt": "Animate the clouds with a slow camera move and natural motion.", "image": { "url": "https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png" }, "duration": 1, "aspect_ratio": "16:9", "resolution": "720p" }; const response = await fetch( "https://api.cometapi.com/grok/v1/videos/generations", { method: "POST", headers: { Authorization: `Bearer ${process.env.COMETAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify(payload), }, ); if (!response.ok) throw new Error(`Request failed: ${response.status}`); console.log((await response.json()).request_id); - lang: Shell label: Image data URI source: |- { printf '%s' '{"model":"grok-imagine-video-1.5","prompt":"Animate the clouds drifting across the sky.","image":{"url":"data:image/png;base64,' base64 < source.png | tr -d '\n' printf '%s' '"},"duration":1,"aspect_ratio":"16:9","resolution":"480p"}' } | curl https://api.cometapi.com/grok/v1/videos/generations \ -H "Authorization: Bearer $COMETAPI_KEY" \ -H "Content-Type: application/json" \ --data-binary @- - lang: Python label: Image data URI source: |- import base64 import os import requests with open("source.png", "rb") as image_file: image_b64 = base64.b64encode(image_file.read()).decode() payload = { "model": "grok-imagine-video-1.5", "prompt": "Animate the clouds drifting across the sky.", "image": { "url": "data:image/png;base64," + image_b64 }, "duration": 1, "aspect_ratio": "16:9", "resolution": "480p" } response = requests.post( "https://api.cometapi.com/grok/v1/videos/generations", headers={ "Authorization": "Bearer " + os.environ["COMETAPI_KEY"], "Content-Type": "application/json", }, json=payload, timeout=180, ) response.raise_for_status() print(response.json()["request_id"]) - lang: JavaScript label: Image data URI source: >- import { readFile } from "node:fs/promises"; const imageB64 = (await readFile("source.png")).toString("base64"); const payload = { "model": "grok-imagine-video-1.5", "prompt": "Animate the clouds drifting across the sky.", "image": { "url": `data:image/png;base64,${imageB64}` }, "duration": 1, "aspect_ratio": "16:9", "resolution": "480p" }; const response = await fetch( "https://api.cometapi.com/grok/v1/videos/generations", { method: "POST", headers: { Authorization: `Bearer ${process.env.COMETAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify(payload), }, ); if (!response.ok) throw new Error(`Request failed: ${response.status}`); console.log((await response.json()).request_id); - lang: Shell label: Reference image source: |- curl https://api.cometapi.com/grok/v1/videos/generations \ -H "Authorization: Bearer $COMETAPI_KEY" \ -H "Content-Type: application/json" \ --data-binary '{ "model": "grok-imagine-video-1.5", "prompt": "Use as the visual reference for a gentle cinematic pan.", "reference_images": [ { "url": "https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png" } ], "duration": 1, "aspect_ratio": "16:9", "resolution": "720p" }' - lang: Python label: Reference image source: |- import os import requests payload = { "model": "grok-imagine-video-1.5", "prompt": "Use as the visual reference for a gentle cinematic pan.", "reference_images": [ { "url": "https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png" } ], "duration": 1, "aspect_ratio": "16:9", "resolution": "720p" } response = requests.post( "https://api.cometapi.com/grok/v1/videos/generations", headers={ "Authorization": "Bearer " + os.environ["COMETAPI_KEY"], "Content-Type": "application/json", }, json=payload, timeout=180, ) response.raise_for_status() print(response.json()["request_id"]) - lang: JavaScript label: Reference image source: >- const payload = { "model": "grok-imagine-video-1.5", "prompt": "Use as the visual reference for a gentle cinematic pan.", "reference_images": [ { "url": "https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png" } ], "duration": 1, "aspect_ratio": "16:9", "resolution": "720p" }; const response = await fetch( "https://api.cometapi.com/grok/v1/videos/generations", { method: "POST", headers: { Authorization: `Bearer ${process.env.COMETAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify(payload), }, ); if (!response.ok) throw new Error(`Request failed: ${response.status}`); console.log((await response.json()).request_id); - lang: Shell label: Reference image and preset voice source: |- curl https://api.cometapi.com/grok/v1/videos/generations \ -H "Authorization: Bearer $COMETAPI_KEY" \ -H "Content-Type: application/json" \ --data-binary '{ "model": "grok-imagine-video-1.5", "prompt": "Use as the visual reference while says Welcome to the observatory.", "reference_images": [ { "url": "https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png" } ], "reference_audios": [ { "voice_id": "eve" } ], "duration": 1, "aspect_ratio": "16:9", "resolution": "720p" }' - lang: Python label: Reference image and preset voice source: |- import os import requests payload = { "model": "grok-imagine-video-1.5", "prompt": "Use as the visual reference while says Welcome to the observatory.", "reference_images": [ { "url": "https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png" } ], "reference_audios": [ { "voice_id": "eve" } ], "duration": 1, "aspect_ratio": "16:9", "resolution": "720p" } response = requests.post( "https://api.cometapi.com/grok/v1/videos/generations", headers={ "Authorization": "Bearer " + os.environ["COMETAPI_KEY"], "Content-Type": "application/json", }, json=payload, timeout=180, ) response.raise_for_status() print(response.json()["request_id"]) - lang: JavaScript label: Reference image and preset voice source: >- const payload = { "model": "grok-imagine-video-1.5", "prompt": "Use as the visual reference while says Welcome to the observatory.", "reference_images": [ { "url": "https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png" } ], "reference_audios": [ { "voice_id": "eve" } ], "duration": 1, "aspect_ratio": "16:9", "resolution": "720p" }; const response = await fetch( "https://api.cometapi.com/grok/v1/videos/generations", { method: "POST", headers: { Authorization: `Bearer ${process.env.COMETAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify(payload), }, ); if (!response.ok) throw new Error(`Request failed: ${response.status}`); console.log((await response.json()).request_id); components: securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: API key description: Use your CometAPI API key as the bearer value. ````