Skip to content

ParthaPRay/Runwayml-Video-Generation-API-Call

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 

Repository files navigation

Runwayml-Video-Generation-API-Call

This repo has codes for runwayml video image generation API call

This code uses DALL-E-3 as base image generator model. User gives text prompt hardcoded into the code to generate an image and download it in local directory called "generated_content". The images are named as image_1.png, image_2.png and so on. Images are downlaoded as .png format with size 1024x1024. The format can be changed to .jpg also.

Then, the image (say image_1.png) is converted to base64 format. Such conerverted format + text prompt together used to call the gen3a_turbo model from runwayml API. The call generated a video of 5 seconds duration. Then the generated video is downloaded as _video_1.mp4 (in .mp4 format) in the same directory i.e. "generated_content". For aother image, nsame file named video is generted.

The code uses one generated image (from DALL-E-3) to generate one video (from runwayml).

Resources

Runwayml API doc is available https://pypi.org/project/runwayml/ and https://docs.dev.runwayml.com/

Buy API credits

Buy API credits for runwayml by making an organization.

Buy ChatGPT4o pro subscription that gives DALL-E-3 service too.

Steps

  1. Install dependencies

    pip install -r requirements.txt

  2. Make a .env file

    The .env file where you replace ENTER-YOUR-DALL-E-API-KEY from chatgpt and ENTER-YOUR-RUNWAYML-API-KEY with your actual API key from runwayml

    OPENAI_API_KEY=ENTER-YOUR-DALL-E-API-KEY

    RUNWAYML_API_SECRET=ENTER-YOUR-RUNWAYML-API-KEY

AND

In a shell terminal (Linux and MAC):

export RUNWAYML_API_SECRET=**ENTER-YOUR-RUNWAYML-API-KEY**

Note: For windows, no key export support exists

  1. Change the "num_generations"

    If num_generations = 2, then 2 images and 2 videos will be generated and downloaded. If num_generations = 5, then 5 images and 5 videos will be generated and downloaded. So, change the value as per your requirement.

  2. Run the code below

    python runwayml_gen3a_turbo.py

This code calls gen3a_turbo image-to-video generation model API to generate 5 second video. This model needs both "Image" + "Text". "Image" is mandatory. The prompt is already given inside the code main_prompt. You can change the main_prompt accordingly.

About

This repo has codes for runwayml video image generation API call

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages