WAN2 Image to Video (I2V) Fully Automated
Just drop in your image and go!
All of the heavy-lifting is done for you. This super convenient workflow takes in an input image, optimally resizes it as needed, runs it through a vision language model, then takes the image description through an LLM to generate a video prompt. Finally, it is translated to Chinese before prompting (as the WAN model is from China and seems to get better results when prompted in Chinese).
Helpful notes are included for changes you may wish to make. Please change the provided example Groq API key as it will be rate limited and cause errors if more than one person uses it as a time.