Stable Diffusion, I2VGen, SVD Technologies

This table is a non-exhaustive list (2025) of tools and models based on diffusion technologies like Stable Diffusion, I2VGen, and SVD, including open source and SaaS options for varied accessibility.

Name Site/Project Description Available Models Temporal Coherence Realism Camera Control Artifacts Accessibility Type Technology Type Resolution / Max Frames / Free Time
Stable Video Diffusion (SVD) Stable Video Diffusion Diffusion model for generating short videos from images, based on Stable Diffusion, with high temporal coherence for 14 to 25 frames. SVD, SVD-XT, SV4D 2.0 ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐ Occasional flickering Free, open source, runnable locally or via Hugging Face Spaces. OSS Project SVD 576x1024 / 25 frames / Unlimited (local)
Stable Video Diffusion 2.0 (SVD 2.0) Stable Video Diffusion 2.0 New version of SVD with support up to 48 frames, better temporal coherence, and native 1080p generation. Integrates an optimized scheduler and motion fine-tuning. SVD 2.0, SVD 2.0-XT, SVD 2.0-Ultra ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐ Sometimes "ghosting" on fast objects Free, open source (via Hugging Face and GitHub) OSS Project SVD 1080p / 48 frames / Unlimited (local)
I2VGen-XL I2VGen-XL Cascade diffusion model for high-quality video synthesis from images, with good generalization across various data types. I2VGen-XL (3.7B parameters) ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Minor morphing Free, open source, via Hugging Face or locally. OSS Project I2VGen 512x512 / 16 frames / Unlimited (local)
AnimateDiff AnimateDiff Extension to animate static images into videos via Stable Diffusion and ComfyUI, by adding motion modules. AnimateDiff v1, v2, v3; Motion Modules (various) ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐ Flickering on fast movements Free, open source, local via ComfyUI. OSS Project Stable Diffusion 512x512 / 16-32 frames / Unlimited (local)
ModelScope i2vgen-XL ModelScope i2vgen-XL Open source image/text-to-video pipeline developed by DAMO Academy, for smooth and coherent generation. i2vgen-XL ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Light distortion Free, open source, via studios or local. OSS Project I2VGen 256x256 / 49 frames / Unlimited (local)
Runway Gen-3 Alpha Runway Gen-3 Alpha SaaS platform to transform images into videos with text and image support, ideal for creative use. Gen-3 Alpha, Gen-4 Turbo ⭐⭐⭐⭐ ⭐⭐⭐⭐⭐ ⭐⭐⭐⭐ Minimal Free trial (125 credits), paid plans from $15/month. Paid SaaS Stable Diffusion 1080p / 16s / 4s (125 credits)
Runway Gen-4 Turbo Runway Gen-4 Turbo Successor to Gen-3 Alpha, with real-time video generation (under 5s), 3D camera support, and enhanced photorealistic fidelity. Compatible with image + text + audio. Gen-4 Turbo ⭐⭐⭐⭐⭐ ⭐⭐⭐⭐⭐ ⭐⭐⭐⭐⭐ Almost none Free trial (125 credits), subscriptions from $15/month Paid SaaS Stable Diffusion 1280x720 / 10s / 10s (125 credits)
Pika Labs Pika Labs "Idea-to-video" tool with image-to-video mode, accessible via web for quick generation. Pika 1.0, Pika 1.5 ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐ Over-stylization Periodic free trials, limited credits. SaaS Stable Diffusion 1080p / 5s / 3s (trial)
Pika 2.0 Pika 2.0 New architecture with joint temporal diffusion and audio-to-motion support. Allows animating an image with an audio track (e.g., lip sync). Pika 2.0, Pika 2.0 Pro ⭐⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐⭐ Over-stylization Early access free, Pro version paid SaaS Stable Diffusion 1080p / 10s / Limited (early access)
Luma Dream Machine Luma Dream Machine Fast image-to-video generator, producing 5-second clips with realistic movements. Dream Machine v1, v2 ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐⭐ Occasional distortion Free access with trials, paid plans. SaaS Stable Diffusion 720p / 120 frames / 5s (trial)
Luma Dream Machine v3 Luma Dream Machine v3 New generation with "motion brush", 4K generation, and <3s latency. Supports image upload + prompt + motion reference. Dream Machine v3 ⭐⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐⭐⭐ Light blur Limited free access (5 videos/day), Pro plans at $24/month SaaS Stable Diffusion 4K / 120 frames / 5s (5 videos/day)
Krea AI Motion Krea AI Motion Tool to transform images into videos by transferring motion, with free functions for Motion/Video. Krea Motion, Video Leap ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐ Moderate morphing Free functions, premium paid. SaaS Stable Diffusion 1080p / 5s / 3s (free)
Pixverse AI Pixverse AI Platform to generate videos from images and text, with a simple web interface. Pixverse v1, v2 ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Frequent flickering Free credits, paid options. SaaS Stable Diffusion 720p / 5s / 3s (free credits)
Vidnoz AI Vidnoz AI AI tool to convert images into videos, with support for various formats and effects. Vidnoz Image-to-Video ⭐⭐ ⭐⭐⭐ ⭐⭐ Visible distortion Free with limits, premium paid. SaaS Stable Diffusion 1080p / 4s / 1 video/day
Hailuo AI Hailuo AI Image-to-video generator focused on HD quality and natural movements. Hailuo MiniMax ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐ Light blur Free trials, paid API. SaaS Stable Diffusion 720p / 5s / 3-5 videos/day
Kling AI Kling AI Chinese AI video model for image-to-video, with high fidelity and camera control. Kling 1.0, 1.5 ⭐⭐⭐⭐ ⭐⭐⭐⭐⭐ ⭐⭐⭐⭐ Light blur in background Limited free credits, paid. SaaS Stable Diffusion 1080p / 5s / 3-4 videos/day
Kling 2.0 Kling 2.0 2025 version of the Chinese platform: 4K resolution, advanced camera control (dolly, zoom, rotation), and generation from 8s to 12s. Kling 2.0, Kling 2.0 Pro ⭐⭐⭐⭐ ⭐⭐⭐⭐⭐ ⭐⭐⭐⭐⭐ Light blur in background Limited free credits, paid API SaaS Stable Diffusion 4K / 10s / 5s (166 free credits)
Wan 2.2 A14B Wan 2.2 A14B Advanced open source model for image-to-video with MoE architecture for better aesthetics and motion. Wan 2.2 I2V A14B ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐ Minor temporal blur Free, open source. OSS Project SVD 1024x576 / 25 frames / Unlimited (local)
Wan 3.0 MoE Wan 3.0 MoE Sparse (Mixture of Experts) model with 12B effective parameters, specialized in natural movements (water, fire, hair). Wan 3.0 I2V MoE ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐ Less good on faces Free, open source OSS Project SVD 1080p / 120 frames / Unlimited (local)
HunyuanVideo HunyuanVideo Open source model from Tencent for video from images, extended in 2025 with i2v capabilities. HunyuanVideo I2V ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐ Moderate ghosting Free, open source. OSS Project I2VGen 720p / 5s / Unlimited (local)
HunyuanVideo I2V v2 HunyuanVideo I2V v2 Major update to the Tencent model with multi-view support, improved temporal interpolation, and generation up to 60 fps. HunyuanVideo-I2V-v2 ⭐⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐⭐ Minimal Free, open source OSS Project I2VGen 720p / 60 fps / Unlimited (local)
Hugging Face Spaces (SVD Apps) Hugging Face Spaces (SVD Apps) Hosted spaces to test SVD and other i2v models, with free Inference API. Various SVD, I2VGen apps ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Varies by app Free API tier, hosted. Hosted Platform SVD Varies / 25 frames / Limited free API
Meta Create/Vibes Meta Create/Vibes Free tool from Meta for image-to-video with lipsync and 16:9/9:16 formats. Vibes, Create ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐ Minimal Unlimited free, no watermark. Free SaaS Stable Diffusion 1080p / 10s / Unlimited
Meta Create v2 (Vibes+) Meta Create v2 (Vibes+) Update to Meta's free tool: image-to-video support with transparent background, camera effects, and 1080p export without watermark. Vibes+, Create v2 ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐⭐ Minimal Free, unlimited, no watermark Free SaaS Stable Diffusion 1080p / 10s / Unlimited
Grok Image-to-Video Grok Image-to-Video xAI functionality to generate videos from images with lipsync, integrable via API. Grok Video ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ API distortion Free via xAI API, fast routes. API/LLM Stable Diffusion 720p / 5s / Limited free API
Haiper Haiper AI platform for video creation from images, focused on creative expression. Haiper I2V ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐ Moderate Free trials, paid. SaaS Stable Diffusion 1080p / 8s / 2 videos/day
Moonvalley Moonvalley Tool for image-to-video with advanced effects, though less practical for some uses. Moonvalley v1 ⭐⭐ ⭐⭐⭐ ⭐⭐ High Limited free, paid. SaaS Stable Diffusion 720p / 5s / 1 video/day
Vheer Vheer Web tool to animate images into videos with camera control and movements. Vheer Motion ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐⭐ Minimal Unlimited free, no watermark. Free SaaS Stable Diffusion 1080p / 5-10s / Unlimited
Qwen Video Qwen Video Model for text-to-video extendable to image-to-video, with effective prompts. Qwen 2.5 Video ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Ghosting Free open source. OSS Project Stable Diffusion 720p / 16 frames / Unlimited (local)
Qwen Video 2.5 Qwen Video 2.5 Improved version with initial image support, 720p/30fps generation, and direct integration into Tongyi Wanxiang. Qwen Video 2.5 ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐ Minimal Open source (Apache 2.0), via ModelScope OSS Project Stable Diffusion 720p / 30 fps / Unlimited (local)
CogVideoX CogVideoX Open source video generation model from text or images, developed by Zhipu AI, with multi-frame support. CogVideoX-2b, CogVideoX-5b ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐ Minor flickering Free, open source, via Hugging Face OSS Project SVD 480p / 49 frames / Unlimited (local)
Open-Sora Open-Sora Open source initiative to reproduce OpenAI's Sora, with image-to-video and text-to-video support. Open-Sora v1.0, v1.1 ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Inconsistent Free, open source OSS Project Stable Diffusion 768x768 / 128 frames / Unlimited (local)
Open-Sora Plan v2 Open-Sora Plan v2 Evolution of Open-Sora with long-duration support (up to 30s), better object movement handling, and custom dataset fine-tuning. Open-Sora-Plan v2.0 ⭐⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐ Moderate on long duration Free, open source OSS Project Stable Diffusion 720p / 30s / Unlimited (local)
Stable Video 3D Stable Video 3D Extension of SVD to generate 3D views from a single image, compatible with NeRF. SV3D, SV3D-u ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐⭐ 3D deformation Free via Hugging Face OSS Project SVD 576x1024 / 25 frames / Unlimited (local)
Emu Video Emu Video Meta's model for video generation from images and text, with high resolution. Emu Video (not open) ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Limited research Restricted access, research Research Stable Diffusion 720p / 2s / N/A (research)
Phenaki Phenaki Google's model for long video from text, adaptable to image-to-video. Phenaki (not open) ⭐⭐ ⭐⭐ ⭐⭐ High Not public Research I2VGen 480p / 10s+ / N/A
Make-A-Video Make-A-Video Meta's ancestor to Emu, for video generation from textual prompts or images. Make-A-Video ⭐⭐ ⭐⭐ N/A Not open Research Stable Diffusion 256x256 / 4s / N/A
VideoCrafter VideoCrafter Open source suite for high-quality video generation from text or static images. VideoCrafter2 ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Moderate Free, open source OSS Project Stable Diffusion 1024x576 / 49 frames / Unlimited (local)
LaVie LaVie Open source model for realistic video generation from text or images. LaVie-base, LaVie-large ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Flickering Free, open source OSS Project Stable Diffusion 512x512 / 16 frames / Unlimited (local)
Tune-A-Video Tune-A-Video Fine-tuning method to adapt Stable Diffusion to personalized video generation. Tune-A-Video ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Depends on fine-tuning Free, open source OSS Project Stable Diffusion 512x512 / 16 frames / Unlimited (local)
Text2Video-Zero Text2Video-Zero Video generation without training, from SD, compatible with initial image. Text2Video-Zero ⭐⭐ ⭐⭐ High Free, open source OSS Project Stable Diffusion 512x512 / 25 frames / Unlimited (local)
ZeroScope ZeroScope Lightweight open source model for video generation from text (adaptable to i2v). ZeroScope v2 ⭐⭐ ⭐⭐ Frequent Free, open source OSS Project Stable Diffusion 1024x576 / 24 frames / Unlimited (local)
ModelScope Text-to-Video ModelScope Text-to-Video DAMO Academy's video suite, including i2v and t2v pipelines. DAMO T2V, I2VGen-XL ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Moderate Free, open source OSS Project I2VGen 256x256 / 49 frames / Unlimited (local)
Stable Diffusion Video (SDV) Stable Diffusion Video (SDV) Library to generate videos from prompts or images with interpolation. SDV v1 ⭐⭐ ⭐⭐ High Free, open source OSS Project Stable Diffusion 512x512 / 16 frames / Unlimited (local)
Deforum Deforum SD prompt animation via interpolation, compatible with initial image. Deforum + SD ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐ Unstable interpolation Free, open source OSS Project Stable Diffusion 512x512 / 100+ frames / Unlimited (local)
AnimateAnyone AnimateAnyone Video generation from image + pose, widely used for human avatars. AnimateAnyone ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Limited pose Free, open source OSS Project Stable Diffusion 512x512 / 16 frames / Unlimited (local)
MagicAnimate MagicAnimate Animation of human images from pose sequences. MagicAnimate ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Human distortion Free, open source OSS Project Stable Diffusion 512x512 / 33 frames / Unlimited (local)
MotionCtrl MotionCtrl Explicit motion control in videos generated from images. MotionCtrl ⭐⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐⭐ Forced motion Free, open source OSS Project SVD 576x1024 / 25 frames / Unlimited (local)
VideoReDo VideoReDo Improvement and reinterpretation of existing videos via diffusion. VideoReDo ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Variable reinterpretation Free, open source OSS Project Stable Diffusion 512x512 / 16 frames / Unlimited (local)
DynamiCrafter DynamiCrafter Video generation from image + prompt, with advanced temporal interpolation. DynamiCrafter ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐ Limited interpolation Free, open source OSS Project Stable Diffusion 512x512 / 17 frames / Unlimited (local)
Rerender Rerender Coherent video generation from static images and prompts. Rerender A ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Variable coherence Free, open source OSS Project Stable Diffusion 512x512 / 16 frames / Unlimited (local)
SVD-XT Turbo SVD-XT Turbo Accelerated version of SVD-XT for fast generation. SVD-XT Turbo ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐ Accelerated causes blur Free, open source OSS Project SVD 576x1024 / 25 frames / Unlimited (local)
Kandinsky Video Kandinsky Video Video extension of the Kandinsky model for image-to-video. Kandinsky-V ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ High artistic Free, open source OSS Project Stable Diffusion 512x512 / 16 frames / Unlimited (local)
CogView3-Video CogView3-Video Video component of CogView3, supports i2v via diffusion. CogView3-V ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Moderate Free, open source OSS Project SVD 480p / 6s / Unlimited (local)
StableCascade Video StableCascade Video Video adaptation of Stable Cascade, compatible with i2v. StableCascade-V ⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐ Cascade blur Free, open source OSS Project Stable Diffusion 1024x1024 / 16 frames / Unlimited (local)
VideoLDM VideoLDM Video generation via Latent Diffusion Models, base for several projects. VideoLDM ⭐⭐ ⭐⭐ High latent Free, open source OSS Project Stable Diffusion 256x256 / 16 frames / Unlimited (local)
Gen-1 (Runway) Gen-1 (Runway) Old Runway model for style transfer video from image. Gen-1 ⭐⭐ ⭐⭐ Deprecated Paid (deprecated) Paid SaaS Stable Diffusion 720p / 4s / N/A (paid)
Gen-2 (Runway) Gen-2 (Runway) Predecessor to Gen-3, supports image-to-video. Gen-2 ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐ Moderate Paid Paid SaaS Stable Diffusion 1080p / 16s / Limited trial
Stable Video Ultra Stable Video Ultra Improved version of SVD not yet released (rumored). SVD-Ultra ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐ N/A Not available Research SVD N/A / N/A / N/A
LTX Studio LTX Studio Web platform to generate videos from images with camera control. LTX-I2V ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐⭐ Moderate Limited free, paid SaaS Stable Diffusion 720p / 5s / Limited free
Infinite Nature Infinite Nature Generation of natural videos from landscape images. Infinite Nature ⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐ Unstable landscape Research Research Stable Diffusion N/A / N/A / N/A
NeRFFaceVideo NeRFFaceVideo Video generation of face from a single image. NeRFFaceVideo ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Deformed face Free, open source OSS Project SVD 256x256 / 100 frames / Unlimited (local)
Vid2Vid Vid2Vid Video-to-video translation, adaptable to i2v via initial image. Vid2Vid ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Variable translation Free, open source OSS Project Stable Diffusion 512x512 / Varies / Unlimited (local)
MoDi MoDi Microsoft's model for multimodal video generation (text + image). MoDi ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Restricted Restricted access Research I2VGen N/A / N/A / N/A
Imagen Video Imagen Video Google's high-resolution video model, not open. Imagen Video ⭐⭐⭐⭐ ⭐⭐⭐⭐ ⭐⭐⭐ N/A Not public Research Stable Diffusion 1024x1024 / 2s / N/A
ComfyUI-I2V Nodes ComfyUI-I2V Nodes Nodes to integrate SVD, AnimateDiff, etc. into ComfyUI. I2V Plugins ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐ Depends on node Free, open source OSS Tool Stable Diffusion Varies / Varies / Unlimited (local)
Stable Diffusion WebUI I2V Stable Diffusion WebUI I2V Extensions for video in WebUI (e.g., SVD plugin). SVD plugin ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Variable plugin Free, open source OSS Tool Stable Diffusion 512x512 / 25 frames / Unlimited (local)
VideoFusion VideoFusion Fusion of multiple images into coherent video. VideoFusion ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Unstable fusion Free, open source OSS Project Stable Diffusion 512x512 / 16 frames / Unlimited (local)
Dreamix Dreamix Video generation from image + motion prompt. Dreamix ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Variable motion Free, open source OSS Project Stable Diffusion 256x256 / 16 frames / Unlimited (local)
CtrlVideo CtrlVideo Spatial and temporal control in video generation. CtrlVideo ⭐⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐⭐ Forced control Free, open source OSS Project Stable Diffusion 512x512 / 16 frames / Unlimited (local)
RAVE RAVE Fast video generation with motion control. RAVE ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐ Fast causes errors Free, open source OSS Project SVD 576x1024 / 14 frames / Unlimited (local)
T2V-Turbo T2V-Turbo Accelerated text-to-video version, adaptable to i2v. T2V-Turbo ⭐⭐ ⭐⭐ High accelerated Free, open source OSS Project Stable Diffusion 512x512 / 24 frames / Unlimited (local)
PixArt Video PixArt Video Video extension of the PixArt model, supports i2v. PixArt-V ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Artistic Free, open source OSS Project Stable Diffusion 1024x1024 / 16 frames / Unlimited (local)
Stable Video 4D Stable Video 4D Generation of 4D videos (time + depth) from images. SV4D 2.0 ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐⭐ 4D deformation Free, open source OSS Project SVD 576x1024 / 25 frames / Unlimited (local)
Vid2Dream Vid2Dream Hugging Face space to transform videos into animated dreams. Vid2Dream ⭐⭐ ⭐⭐ Unstable dream Free Hosted Platform Stable Diffusion 512x512 / 16 frames / Limited API
MotionBrush MotionBrush Tool to animate specific areas of an image. MotionBrush ⭐⭐⭐ ⭐⭐⭐ ⭐⭐ Limited zone Limited free SaaS Stable Diffusion 720p / 5s / Limited free
DeepMotion Animate 3D DeepMotion Animate 3D 3D video generation from 2D images. Animate 3D ⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐ 3D conversion Freemium SaaS Stable Diffusion 720p / 10s / Limited freemium