A 1.3B parameter text-to-video model supporting multi-resolution training, capable of predicting first and last frames

MultimodalDiffusersMultiple Languages

Eagle 2.5 is a cutting-edge vision-language model (VLM) designed for long-context multimodal learning, supporting the processing of video sequences up to 512 frames and high-resolution images.

MultimodalTransformersOther

A multimodal video-text model built upon InternVideo2-1B and Qwen2.5-7B, using only 16 tokens per frame and supporting input sequences of up to 10,000 frames.

MultimodalTransformersEnglish

Step-Video-T2V is a cutting-edge text-to-video pre-trained model with 30 billion parameters, capable of generating videos up to 204 frames in length.

MultimodalDiffusers

A novel video diffusion model capable of generating high-quality videos from any number of context frames

LoRA adapter trained based on the Flex.1-alpha model, specializing in generating high-quality background images

MultimodalDiffusers

A LoRA model trained on the Flex.1-alpha base model, specializing in generating magnificent natural landscape images, particularly excelling in sunset mountain scenes

Computer VisionDiffusers

VideoChat-Flash-7B is a multimodal model built upon UMT-L (300M) and Qwen2-7B, using only 16 tokens per frame and supporting input sequences of up to approximately 10,000 frames.

MultimodalTransformersEnglish

The NVIDIA Low Frame Rate Speech Codec is a neural audio codec that utilizes limited scalar quantization and adversarial training techniques with large speech language models to achieve high-quality audio compression at a bit rate of 1.89 kbps and 21.5 frames per second, supporting multiple languages.

Audio ProcessingNemo

CogVideoX1.5-5B is a text-to-video generation model open-sourced by Zhipu AI, which can generate high-quality videos based on English text descriptions. This model supports generating videos with a resolution of 1360×768 and a maximum of 81 frames (5 seconds), and has wide application value in the field of video creation.

MultimodalDiffusersEnglish

A zero-shot video editing solution based on diffusion models, improving temporal coherence and reducing memory consumption by merging self-attention tokens across video frames.

MultimodalDiffusers

A model for classifying 20-millisecond audio frames to detect filler pauses (e.g., 'eee', 'errm', etc.)

Audio ProcessingSafetensorsOther

AnimateDiff is a method that utilizes existing Stable Diffusion text-to-image models to create videos by inserting motion module layers to achieve coherent motion between frames.

MultimodalDiffusers

VidXTend is a streamlined encapsulation of StreamingT2V's second-stage video extension process, designed to extend 16-frame 256x256 pixel animations by 8 frames each time (equivalent to 1 second duration at 8fps)

Computer VisionDiffusers

Stable Video Diffusion (SVD) 1.1 is a diffusion model-based image-to-video tool capable of generating short video clips from static images as conditional frames.

Computer VisionDiffusers

AnimateDiff is a technology that leverages existing Stable Diffusion text-to-image models to create videos by inserting motion module layers to achieve coherent motion between image frames.

Computer VisionDiffusers

A diffusion model that generates short video clips using static images as conditional frames

Computer VisionDiffusers

AnimateDiff is a technology that enables existing Stable Diffusion text-to-image models to generate videos by inserting motion module layers to achieve coherent motion between frames.

Computer VisionDiffusers

Hotshot-XL is an AI text-to-GIF model based on Stable Diffusion XL, capable of generating 1-second GIF animations at 8 frames per second.

MultimodalDiffusers

A Stable-Diffusion-xl version fine-tuned with 125 comic frames from 'The Adventures of Tintin', capable of generating classic characters like Tintin, Snowy (Milou), and Captain Haddock.

MultimodalDiffusers