How does Conch AI Video integrate into ComfyUI nodes?

To better use Conch AI (MiniMax) video generation for the image domain, we maintain a ComfyUI node.

ComfyUI MiniMax Video is a powerful extension tool that integrates the image-to-video and text-to-video generation features of MiniMax AI. Users can easily convert static images to dynamic video, support multiple video generation models, and provide real-time video preview and custom cue optimization. The tool's flexible API configuration makes it suitable for a variety of video generation needs, especially for users who need to quickly generate high-quality video content.

Comfyui Access API

1. Prepare comfyui environment, go to https://github.com/comfyanonymous/ComfyUI下载免安装文件, unzip it, click run_nvidia_gpu.bat to start comfyui if you have a graphics card, click run_cpu.bat to start if you don't have a graphics card, keep it running in the background after startup and don't close it. If you have a graphics card, click run_cpu.bat to start comfyui.
2. Download the MiniMax video node files into the cutom_nodes folder: git hub download https://github.com/backearth1/ComfyUI-MiniMax-Video.git
3. Restart comfyui to open the video to generate the workflow file

caveat

1. The video will be downloaded to comfyui's default output directory and named with "time+trace_id".
2. Access address, https://api.minimax.chat/v1 for domestic users; https://api.minimaxi.chat/v1 for overseas users, please note that the corresponding api_key is used.
3. Without the image node, it becomes a Vincentian video t2v
4. The video preview box needs to be adjusted manually, and will be optimized later on
5. During the generation process, you can observe the progress of the generation in the comfyui runtime.
6. Suggest installing another comfyui manager yourself to maintain various new nodes https://github.com/ltdrdata/ComfyUI-Manager