Paddle Multimodal Integration and eXploration, supporting mainstream multi-modal tasks, including end-to-end large-scale multi-modal pretrain models and diffusion model toolbox. Equipped with high performance and flexibility.
image-to-text clip text-to-image dit multimodal sora text-to-video aigc stable-diffusion controlnet llava sd-xl ppdiffusers eva-clip stablevideodiffusion minicpm-v internvl2 qwen2-vl got-ocr20 deepseek-vl
-
Updated
Mar 6, 2026 - Python