POST
v1
/
replicate
/
videogeneration
Video Generation
curl --request POST \
  --url http://localhost:9000/api/v1/replicate/videogeneration \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: multipart/form-data' \
  --form 'prompt=<string>' \
  --form 'negative_prompt=<string>' \
  --form init_weight=123 \
  --form num_frames=123 \
  --form num_inferences_steps=123 \
  --form width=123 \
  --form height=123 \
  --form guidance_scale=123 \
  --form fps=123 \
  --form 'video_model=<string>' \
  --form batch_size=123 \
  --form remove_watermark=true

​
Model

​
Query

​
model
string
default:"anotherjesse/zeroscope-v2-xl"
required
The image model to be used;

​
Body

​
prompt
string
required
Input prompt
​
negative_prompt
string
default:"blurred, noisy, washed out, distorted, broken"
Factor to scale image by
​
init_weight
float
default:"0.5"
URL of the initial video (optional)
​
num_frames
int
default:"24"
Number of frames for the output video
​
num_inferences_steps
int
default:"50"
Number of denoising steps
​
width
int
default:"1024"
Width of the output video
​
height
int
default:"1024"
Height of the output video
​
guidance_scale
float
default:"17.5"
Guidance scale
​
fps
int
default:"10"
fps for the output video
​
video_model
string
default:"xl"
Model to use
​
batch_size
int
default:"1"
Batch size
​
remove_watermark
boolean
default:"false"
Remove watermark