Wan2.2 is a significant upgrade version of the basic video model. It introduces the Mixture of Experts (MoE) architecture, incorporates carefully curated aesthetic data, and is trained on larger datasets to enhance the ability to generate complex motions. This model supports generating 5-second videos with 480P and 720P resolutions, and has significant improvements in video generation quality and performance.

MultimodalDiffusers

Wan2.2-S2V-14B is a Mixture of Experts (MoE) model designed for audio-driven cinema-level video generation. It can generate high-quality video content based on input audio, reference images, and text prompts, supports 480P and 720P resolutions, and features complex motion generation and cinema-level aesthetic effects.

MultimodalDiffusersSpanish

This is the DFloat11 lossless compressed version of the Wan-AI/Wan2.2-T2V-A14B model. Through advanced compression technology, the model size is reduced by 32% while maintaining bit-identical output quality. It supports efficient 720P video generation on a single 24GB GPU.

MultimodalDiffusers

SkyReels-V2 is a 14B-parameter text-to-video generation model that supports 720P resolution output and includes VACE functionality.

MultimodalGgufEnglish

SkyReels-V2-DF-14B-720P is a video generation model that supports generating 720P resolution videos from images.

Computer VisionGguf

SkyReels V2 is an unlimited-length cinematic generation model that adopts an autoregressive diffusion force architecture, supports 720P HD video generation, and achieves SOTA performance among public models.

Computer VisionSafetensors

Open-source image-to-video generation model supporting 360p to 720p resolution with up to 5-second video generation

Computer VisionDiffusersEnglish

EAGLE is a decoder causal language model with 3.1 billion parameters developed by the Electronics and Telecommunications Research Institute (ETRI) in South Korea, focusing on the STEM fields, especially mathematics and quantitative reasoning. This model has been pre-trained on 720 billion tokens and uses a LLaMA-compatible architecture. However, it is not a general-purpose language model and requires additional fine-tuning to be used in applications such as chatbots.

Natural Language ProcessingTransformersKorean

Allegro is an open-source high-quality text-to-video generation model capable of producing 6-second detailed videos at 720x1280 resolution and 15 FPS.

MultimodalDiffusersEnglish

CogVideoX-5B-I2V is a high-quality image-to-video generation model open-sourced by the Qingying platform. It can generate a 6-second dynamic video based on the input static image, with a resolution of 720×480 and a frame rate of 8fps.

MultimodalDiffusersEnglish

CogVideoX-5B is an open-source video generation model derived from QingYing. It can generate high-quality videos based on text input, providing powerful support for video creation. This model has 5 billion parameters and can generate 720x480 resolution videos with a duration of 6 seconds.

MultimodalDiffusersEnglish

This is an unconditional diffusion model for generating lung ultrasound images, capable of producing medical images at 720x720 resolution.

Computer VisionDiffusers