Skip to main content
POST
/
v1
/
video
POST /v1/video — Generate Videos via API
curl --request POST \
  --url https://api.example.com/v1/video \
  --header 'Content-Type: application/json' \
  --data '
{
  "prompt": "<string>",
  "model": "<string>",
  "mode": "<string>",
  "image": "<string>",
  "aspect_ratio": "<string>",
  "duration": 123
}
'
{
  "request_id": "<string>",
  "status": "<string>",
  "video_url": "<string>",
  "credits_used": 123
}

Documentation Index

Fetch the complete documentation index at: https://docs.artnex.app/llms.txt

Use this file to discover all available pages before exploring further.

The /v1/video endpoint generates video clips from text prompts or a reference start image. Because video generation is computationally intensive, this endpoint is asynchronous — your initial request returns a request_id, and you poll the status endpoint until the video is ready.

​
Endpoint

POST https://api.artnex.ai/v1/video

​
Request parameters

​
prompt
string
required
Text description of the video you want to generate. Describe the scene, motion, style, and any camera movements. Maximum 4,000 characters.Example: "A drone flying slowly over a misty mountain range at sunrise, cinematic, 4K"
​
model
string
required
The video model to use. See the Models table for all supported model IDs and their credit costs.
​
mode
string
default:"text-to-video"
Generation mode. Use "text-to-video" to generate from a prompt alone, or "image-to-video" to animate a reference image. When using "image-to-video", the image field is required.
​
image
string
Start image for image-to-video generation. Required when mode is "image-to-video". Provide a publicly accessible HTTPS URL or a base64-encoded image string.
​
aspect_ratio
string
default:"16:9"
Output aspect ratio. Accepted values: "16:9", "9:16", "1:1". Not all models support all ratios; unsupported values fall back to the model’s default.
​
duration
integer
Clip duration in seconds. Accepted values depend on the model (typically 4–12). Refer to the model’s supported durations in the Models table.

​
Response fields

​
request_id
string
required
Unique identifier for your generation job. Use this to poll the status endpoint.
​
status
string
Initial job status. Typically "pending" for new requests.
​
video_url
string
Direct HTTPS URL to the generated video. This field is only present when status is "completed". Videos are available for 24 hours.
​
credits_used
integer
Credits deducted from your account when the job completes.

​
Async polling

Video generation takes between 30 seconds and several minutes depending on the model and clip length. After you receive a request_id, poll the status endpoint at regular intervals: 
GET https://api.artnex.ai/v1/video/status/{request_id}
Poll every 5 seconds until status is "completed" or "error". Do not poll more frequently than once per 5 seconds or your requests may be throttled.
Set a maximum polling timeout of 5 minutes (60 attempts at 5-second intervals). If a job does not complete within that window, treat it as failed and retry.

​
Models

​
Text-to-video models

Model IDCreditsSupported durationsAspect ratios
veo3.13,0004, 6, 8 s16:9, 9:16
veo3-fast2,0004, 6, 8 s16:9, 9:16
sora-23,5004, 8, 12 s16:9, 9:16
sora-2-pro6,0004, 8, 12 s16:9, 9:16
kling-v2.5-turbo-pro1,5005, 10 s16:9, 9:16, 1:1
kling-v2.6-pro5,5005, 10, 16 s16:9, 9:16, 1:1
wan-2.53,5005, 10 s16:9, 9:16, 1:1
seedance-v1.5-pro-t2v2,6004–12 s16:9, 9:16, 1:1

​
Image-to-video models

Model IDCreditsSupported durations
veo3.1-i2v3,5004, 6, 8 s
veo3-fast-i2v2,0004, 6, 8 s
kling-v2.5-turbo-pro-i2v1,5005, 10 s
wan-2.5-i2v1,5005–10 s
seedance-v1.5-pro-i2v1,5004–12 s

​
Example requests

# Text-to-video
curl --request POST \
  --url https://api.artnex.ai/v1/video \
  --header "Authorization: Bearer YOUR_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "prompt": "A drone flying slowly over a misty mountain range at sunrise, cinematic, 4K",
    "model": "veo3-fast",
    "mode": "text-to-video",
    "aspect_ratio": "16:9",
    "duration": 6
  }'

​
Example responses

Initial response (job queued): 
{
  "request_id": "req_vid_abc123xyz",
  "status": "pending",
  "credits_used": 0
}
Status response (completed): 
{
  "request_id": "req_vid_abc123xyz",
  "status": "completed",
  "video_url": "https://cdn.artnex.ai/video/req_vid_abc123xyz/output.mp4",
  "credits_used": 2000
}

​
Error codes

HTTP statusError codeCause
400missing_promptThe prompt field is absent or empty
400invalid_modelThe model value is not a supported video model ID
400invalid_modemode must be "text-to-video" or "image-to-video"
400missing_imagemode is "image-to-video" but no image was provided
401invalid_api_keyThe API key is missing, malformed, or revoked
402insufficient_creditsYour account does not have enough credits
429rate_limit_exceededYou have exceeded your plan’s daily request quota
Use veo3-fast (2,000 credits) or kling-v2.5-turbo-pro (1,500 credits) for iteration. Switch to veo3.1 or sora-2 for final production clips.