DynamiCrafter
DynamiCrafter is an open-source image animation model developed by Tencent that brings still images to life by leveraging video diffusion priors to generate dynamic content with natural motion and temporal coherence. Released in October 2023, DynamiCrafter addresses open-domain image animation, where the model must infer plausible motion from a single static image without additional motion guidance. Built on a 1.4 billion parameter diffusion architecture, it utilizes a pre-trained video diffusion model as a motion prior, conditioning generation on the input image to produce animations maintaining the source's visual characteristics while introducing contextually appropriate temporal dynamics. The architecture combines image understanding with learned motion patterns, enabling animation of diverse content including landscapes with moving elements, portraits with subtle expressions, architectural scenes, and artistic compositions. DynamiCrafter demonstrates particular strength in generating physically plausible animations respecting spatial layout and depth relationships, avoiding warping distortions and unnatural deformations. The model supports multiple resolutions and varying animation lengths for different creative and commercial applications. Key use cases include animated photographs for social media, dynamic backgrounds for presentations, bringing artwork to life, and producing visual effects for creative projects. Available under the Apache 2.0 license, DynamiCrafter is accessible on Hugging Face, Replicate, and fal.ai, with community adoption through popular creative workflows. The model represents an important advancement in unsupervised image animation, offering a practical solution for content creators who need to add motion to static visual assets without manual animation skills.
Подробная версия доступна на английском языке.
ElevenLabs Turbo v2.5
ElevenLabs Turbo v2.5 is the fastest commercial text-to-speech model developed by ElevenLabs, specifically optimized for real-time applications requiring minimal latency between text input and audio output. Built on a proprietary architecture, the model delivers near-instantaneous speech synthesis with latencies as low as 300 milliseconds, making it suitable for live conversational AI agents, interactive voice response systems, and real-time translation services. Despite its focus on speed, Turbo v2.5 maintains remarkably natural and expressive speech quality with appropriate prosody, breathing patterns, and emotional nuance. The model supports 32 languages with native-quality pronunciation and can leverage ElevenLabs' voice cloning technology to speak in custom cloned voices, professional voice library voices, or synthetic designer voices. Turbo v2.5 is available exclusively through ElevenLabs' cloud API as a proprietary service with usage-based pricing tiers ranging from a free tier for experimentation to enterprise plans for high-volume production use. The API provides simple integration through REST endpoints and official SDKs for Python, JavaScript, and other popular languages. Key applications include powering AI chatbots and virtual assistants with voice output, creating real-time dubbed content, building accessible applications that convert text to speech on the fly, automated customer service systems, gaming NPC dialogue, and live streaming tools. The model handles SSML tags for fine-grained control over pronunciation, pauses, and emphasis, and supports streaming audio output for immediate playback as generation progresses.
Подробная версия доступна на английском языке.
Era3D
Era3D is a multi-view generation model developed by Alibaba that produces high-resolution, camera-aware multi-view images and normal maps from single input images for 3D reconstruction. The model introduces two key innovations that address common limitations in multi-view generation: a focal length estimation module that adapts to the camera perspective of the input image, and an efficient row-wise attention mechanism that enables generation at higher resolutions than competing methods while using less GPU memory. Era3D generates six consistent views along with corresponding normal maps at 512x512 resolution, providing rich geometric information for downstream 3D mesh reconstruction. The camera-aware design means the model can handle input images taken from different perspectives and focal lengths without degradation in output quality, a significant improvement over methods that assume a fixed camera model. The row-wise attention mechanism replaces the computationally expensive full cross-view attention with a more efficient alternative that processes attention along horizontal rows, reducing memory requirements while maintaining view consistency. Released in May 2024 under the Apache 2.0 license, Era3D is fully open source with code and pre-trained weights available on GitHub. The model demonstrates strong performance across diverse object categories and produces clean multi-view outputs suitable for high-quality 3D reconstruction. Era3D is particularly valuable for professional 3D content creation workflows where input images come from varied sources with different camera characteristics, and where high-resolution multi-view generation is essential for capturing fine details in the final 3D models.
Подробная версия доступна на английском языке.
F5-TTS
F5-TTS is an open-source text-to-speech model developed by SWivid that achieves fast and high-quality speech synthesis through a novel flow matching approach. The model uses a non-autoregressive architecture based on flow matching, learning smooth transformation paths between noise and target speech distributions, enabling efficient single-pass generation significantly faster than autoregressive TTS methods while maintaining comparable quality. F5-TTS supports voice cloning from short reference audio, allowing speech generation in a target speaker's voice from just a few seconds of sample audio. It reproduces vocal characteristics including timbre, pitch range, speaking rhythm, and accent with notable accuracy. A key advantage is inference speed, delivering real-time or faster-than-real-time synthesis on modern GPUs, suitable for interactive and latency-sensitive applications. The model generates speech with natural prosody, appropriate emotional expression, and contextually aware pausing and emphasis patterns. F5-TTS handles multiple languages and produces output at high sample rates suitable for professional audio production. The architecture's simplicity compared to complex multi-stage TTS pipelines makes it easier to train, fine-tune, and deploy in production environments. Released under an open-source license, F5-TTS provides a free alternative to commercial TTS services for research and production use cases. Common applications include voiceover generation, audiobook narration, accessibility tools, virtual assistant voices, podcast production, and automated voice generation for applications requiring personalized speech. Available through Hugging Face with Python integration and ONNX export for cross-platform deployment.
Подробная версия доступна на английском языке.
FidelityFx Super Resolution
FidelityFX Super Resolution (FSR) is AMD's open-source spatial upscaling technology designed to boost performance in real-time rendering applications, particularly video games. Unlike NVIDIA's DLSS which requires dedicated Tensor Cores, FSR is hardware-agnostic and runs on AMD, NVIDIA, and Intel GPUs including integrated graphics. The technology has evolved through multiple generations: FSR 1.0 used Lanczos-based spatial upscaling on single frames, FSR 2.0 introduced temporal upscaling leveraging motion vectors and previous frame data for near-native quality, and FSR 3.0 added optical flow-based frame generation to dramatically increase perceived frame rates. Quality modes range from Ultra Quality to Ultra Performance, letting users balance visual fidelity against performance gains of up to 2x or more. FSR supports DirectX 11, DirectX 12, and Vulkan APIs and is deployed across PC, Xbox, PlayStation, and portable devices like Steam Deck where it enables playable frame rates within limited GPU power budgets. Hundreds of major titles including Cyberpunk 2077, Starfield, and Hogwarts Legacy feature FSR integration, with engine-level support in Unreal Engine and Unity simplifying adoption. Released under the MIT license through AMD's GPUOpen platform, FSR encourages transparent collaboration and modification by developers and researchers. Its platform independence and open-source nature have made it one of the most widely adopted upscaling solutions in the gaming industry, shaping the future of real-time image quality enhancement.
Подробная версия доступна на английском языке.
FLUX Fill
FLUX Fill is the specialized inpainting and outpainting model within the FLUX model family developed by Black Forest Labs, designed for professional-grade region editing, content filling, and image extension. Built on the 12-billion parameter Diffusion Transformer architecture that powers all FLUX models, FLUX Fill takes an input image along with a binary mask indicating the region to be modified and generates seamlessly blended content that matches the surrounding context in style, lighting, perspective, and detail level. The model excels at both inpainting tasks where masked areas within an image are filled with contextually appropriate content and outpainting tasks where image boundaries are extended to create larger compositions. FLUX Fill leverages the superior prompt adherence of the FLUX architecture, allowing users to guide the generation with text descriptions of what should appear in the masked region, providing precise creative control over the output. The model handles complex scenarios including filling regions that span multiple materials and textures, maintaining structural continuity of architectural elements, and generating photorealistic human features in masked face areas. As a proprietary model, FLUX Fill is accessible through Black Forest Labs' API and partner platforms including Replicate and fal.ai, with usage-based pricing. Professional photographers use FLUX Fill for removing unwanted elements and extending compositions, e-commerce teams employ it for product background replacement, digital artists leverage it for creative compositing, and marketing professionals use it for adapting images to different aspect ratios and formats without losing content quality.
Подробная версия доступна на английском языке.
FLUX LoRA
FLUX LoRA is a comprehensive fine-tuning framework and adapter ecosystem built around the LoRA (Low-Rank Adaptation) technique for customizing FLUX image generation models with custom styles, subjects, and concepts. LoRA adapters with typically 1 to 50 million parameters inject trainable low-rank matrices into the attention layers of the base FLUX model, enabling efficient specialization without modifying the original 12-billion parameter weights. This approach dramatically reduces the computational requirements for customization, allowing users to train custom LoRA adapters on consumer GPUs with as little as 8GB VRAM using just 15 to 30 training images in under an hour. The resulting adapter files are compact, typically between 50 and 200 megabytes, and can be loaded on top of any FLUX base model at inference time to activate the learned style or subject. The FLUX LoRA ecosystem has grown rapidly with thousands of community-created adapters available on platforms like CivitAI and Hugging Face, covering diverse styles from photorealistic portraits and anime to specific artistic techniques, brand identities, and individual face or product appearances. Multiple LoRA adapters can be combined simultaneously with adjustable weights, enabling creative blending of different styles and concepts. Released under the Apache 2.0 license, the training tools are fully open source and integrate with popular platforms including the Diffusers library, kohya-ss trainer, ai-toolkit, and ComfyUI. Key applications include creating brand-consistent visual identities, training product-specific models for e-commerce, developing custom artistic styles, generating consistent character appearances across multiple images, and personalizing AI image generation for individual creative workflows.
Подробная версия доступна на английском языке.
FLUX Redux
FLUX Redux is the specialized image variation model within the FLUX model family developed by Black Forest Labs, designed for generating creative variations of reference images while preserving their core style, color palette, and compositional essence. Built on the 12-billion parameter Diffusion Transformer architecture, FLUX Redux takes a reference image as input and produces new images that maintain the visual DNA of the original while introducing controlled variations in content, composition, or perspective. The model captures high-level stylistic attributes including artistic technique, color harmony, lighting mood, and textural qualities, then applies them to generate fresh compositions that feel aesthetically consistent with the source material. FLUX Redux can be combined with text prompts to guide the direction of variation, allowing users to request specific changes like 'same style but with a mountain landscape' or 'similar color palette with an urban scene.' This makes it particularly powerful for brand consistency workflows where marketing teams need multiple visuals sharing a unified aesthetic. The model also supports image-to-image workflows where the reference serves as a strong stylistic prior while text prompts define new content. As a proprietary model, FLUX Redux is accessible through Black Forest Labs' API and partner platforms including Replicate and fal.ai with usage-based pricing. Key applications include generating cohesive visual content series for social media campaigns, creating style-consistent variations for A/B testing in advertising, producing product imagery in consistent brand aesthetics, and creative exploration where artists iterate on a visual direction without starting from scratch.
Подробная версия доступна на английском языке.
FLUX.1 [dev]
Black Forest Labs’ 12B text-to-image model with downloadable weights under a non-commercial model license. Generated images have separate usage terms.
Подробная версия доступна на английском языке.
FLUX.1 [schnell]
FLUX.1 [schnell] is the fastest variant in the FLUX.1 model family, engineered by Black Forest Labs specifically for near real-time image generation. The model achieves remarkable speed by requiring only 1 to 4 inference steps compared to the 28 steps needed by FLUX.1 [dev], making it ideal for interactive applications, live previews, and rapid prototyping workflows. Built on the same Flow Matching architecture as its siblings but optimized through aggressive step distillation, Schnell maintains surprisingly high image quality despite its dramatic speed advantage. The model generates images in under one second on modern GPUs, enabling use cases that were previously impractical with diffusion models such as real-time creative tools and responsive design assistants. Released under the Apache 2.0 open-source license, FLUX.1 [schnell] is freely available for both personal and commercial use. It supports the same 12-billion parameter architecture and can be run locally with 12GB or more VRAM or accessed through cloud APIs on Replicate, fal.ai, and Together AI. The model integrates with ComfyUI and the Diffusers library for flexible deployment. While it trades some fine detail and complex scene accuracy compared to the dev and pro variants, its speed-to-quality ratio is unmatched in the open-source ecosystem. Game developers, UI designers, and application developers building AI-powered creative tools particularly benefit from Schnell's instant generation capability.
Подробная версия доступна на английском языке.
FLUX.1 Kontext [pro]
FLUX.1 Kontext [pro] is a Black Forest Labs model for text-to-image generation and editing with a single reference image. It is served through an API; downloadable weights of the separate Dev variant have a different license.
Подробная версия доступна на английском языке.
FLUX.1 LoRA
FLUX.1 LoRA is the Low-Rank Adaptation fine-tuning framework for the FLUX.1 model family, enabling users to customize the powerful 12-billion parameter FLUX.1 models with their own training data to create specialized image generation models. LoRA works by adding small trainable adapter layers to the frozen base model weights, allowing efficient fine-tuning that captures specific styles, characters, objects, or visual concepts without requiring the computational resources needed for full model training. With FLUX.1 LoRA, users can train custom models using as few as 15 to 30 reference images, making personalized AI image generation accessible to individual creators and small teams. The resulting LoRA adapters are compact files typically ranging from 50MB to 200MB that can be loaded on top of any compatible FLUX.1 base model at inference time. Common use cases include training consistent character representations, brand-specific visual styles, product appearance models, specific artistic techniques, and custom aesthetic preferences. The FLUX.1 LoRA ecosystem has grown rapidly, with thousands of community-created LoRAs available on platforms like CivitAI and Hugging Face covering diverse styles from anime characters to photographic presets. Training can be performed using tools like kohya-ss, ai-toolkit, and various cloud-based training platforms. LoRA models are compatible with ComfyUI, the Diffusers library, and other FLUX.1-supporting interfaces. Professional designers, brand managers, game studios, and content creators requiring consistent visual identity across generated images particularly benefit from FLUX.1 LoRA's customization capabilities.
Подробная версия доступна на английском языке.
FLUX1.1 [pro] Ultra
FLUX1.1 [pro] Ultra is a Black Forest Labs image generation model served through an API. It supports output up to 4 megapixels, Raw mode, and image references.
Подробная версия доступна на английском языке.
Gemini 2.0 Flash
Gemini 2.0 Flash is Google DeepMind's latest multimodal AI model optimized for speed, efficiency, and native multimodal output including text, images, and audio. Released in December 2024, it is the first model in the Gemini family to support native image generation alongside its strong reasoning, coding, and language capabilities. Gemini 2.0 Flash can generate and edit images within conversational context, create visual content from text descriptions, and combine text and image outputs in a single response. The model processes text, images, video, and audio inputs, making it one of the most versatile multimodal models available. For design-related tasks, Gemini 2.0 Flash can generate illustrations, diagrams, infographics, and visual concepts while maintaining conversational context for iterative refinement. The model is notably faster than Gemini 1.5 Pro while matching or exceeding its quality on most benchmarks. Available through Google AI Studio, the Gemini API, and integrated into Google products including Gemini Advanced, the model serves developers, creative professionals, and enterprise users. Gemini 2.0 Flash supports a 1 million token context window, enabling processing of extensive documents, codebases, and multimedia content. The model includes Google's AI safety features and SynthID watermarking for generated images.
Подробная версия доступна на английском языке.
Gemini 3.5 Flash
Gemini 3.5 Flash is Google DeepMind's fastest frontier AI model, designed specifically for agentic tasks, coding, and long-horizon reasoning. Announced at Google I/O 2026, it delivers 289 tokens per second — four times faster than comparable frontier models — while scoring within two points of Anthropic's flagship on coding benchmarks at one-third the cost. The model achieves 76.2% on Terminal-Bench 2.1, 1656 Elo on GDPval-AA, 83.6% on MCP Atlas (tool use), and 84.2% on CharXiv Reasoning (multimodal). It introduces a multi-agent architecture with Antigravity harness for parallel subagent execution, making it exceptionally capable at complex, multi-step software engineering tasks. With a 1 million token context window, it handles entire codebases, long documents, and extended conversations without context degradation. Priced at $1.50 per million input tokens and $9.00 per million output tokens, it offers frontier-level intelligence at Flash-tier pricing. Available through Google AI Studio and the Gemini API with general availability from day one.
Подробная версия доступна на английском языке.
GFPGAN
GFPGAN is a practical face restoration algorithm developed by Tencent ARC that leverages generative facial priors embedded in a pre-trained StyleGAN2 model to restore severely degraded face images with remarkable quality. First released in December 2021, GFPGAN addresses the challenging problem of blind face restoration where input images may suffer from unknown combinations of low resolution, blur, noise, compression artifacts, and other forms of degradation. The model's architecture combines a degradation removal module with a StyleGAN2-based generative prior, using a novel channel-split spatial feature transform layer that balances fidelity to the original face with the high-quality facial details provided by the generative model. This approach allows GFPGAN to restore fine facial details including skin textures, eye clarity, hair strands, and tooth definition that are completely lost in the degraded input. The model processes faces through a U-Net encoder that extracts multi-resolution features from the degraded image, which then modulate the StyleGAN2 decoder's feature maps to produce a restored output that preserves the original identity while dramatically enhancing quality. GFPGAN excels in old photo restoration, enhancing low-resolution surveillance footage, improving video call quality, recovering damaged family photographs, and preparing low-quality source material for professional use. The model is open source under Apache 2.0, available on Hugging Face and Replicate, and has become a foundational component integrated into numerous creative AI tools and pipelines. Its ability to handle real-world degradation patterns rather than just synthetic corruption makes it particularly valuable for practical restoration tasks encountered by photographers, archivists, and content creators.
Подробная версия доступна на английском языке.
GPT-4o Image Generation
GPT-4o Image Generation is OpenAI's natively multimodal image generation capability integrated directly into the GPT-4o model, released in March 2025. Unlike DALL-E which operates as a separate image model, GPT-4o generates images as part of its core reasoning process, enabling unprecedented text rendering accuracy, precise instruction following, and seamless multi-turn image editing through conversation. The model excels at generating images with perfectly readable text, accurate typography, and complex layouts that previous models struggled with. GPT-4o can iterate on generated images through natural language conversation, allowing users to request specific changes like moving elements, adjusting colors, modifying text, or changing styles without starting over. The model demonstrates remarkable understanding of design concepts, branding requirements, and visual communication principles. It supports photorealistic images, illustrations, diagrams, infographics, memes, charts, and technical drawings with consistent quality. The model maintains character and style consistency across multiple generations in the same conversation, enabling coherent visual storytelling and brand content creation. Available through ChatGPT (Plus, Team, and Enterprise) and the OpenAI API, GPT-4o image generation has rapidly become one of the most popular AI image creation tools due to its conversational editing approach and superior text handling. The model includes safety measures and C2PA Content Credentials metadata for AI content identification.
Подробная версия доступна на английском языке.
GPT-Image-2.5 Flare
GPT-Image-2.5 Flare creates and edits images from text and visual references. Evaluate product fidelity, revision effort and API costs before using it for campaign delivery.
Подробная версия доступна на английском языке.
GroundingDINO
Grounding DINO is a powerful open-set object detection model developed by IDEA Research that locates and identifies any object in an image based on natural language text descriptions, representing a paradigm shift from fixed-category detection to language-guided visual understanding. With 172 million parameters, the model combines the DINO detection architecture with text grounding capabilities, enabling it to detect objects that were never seen during training simply by describing them in words. Unlike traditional object detectors trained on fixed categories like COCO's 80 classes, Grounding DINO can find arbitrary objects, parts, materials, or visual concepts by accepting free-form text queries such as 'red shoes on the shelf' or 'cracked window in the building.' The architecture fuses visual features from the image encoder with textual features from a text encoder through cross-modality attention layers, learning to align visual regions with their semantic descriptions. Grounding DINO achieves state-of-the-art results on zero-shot object detection benchmarks and when combined with SAM (Segment Anything Model) creates a powerful pipeline for text-prompted segmentation of any visual concept. Released under the Apache 2.0 license, the model is fully open source and widely used in computer vision research and production systems. Key applications include automated image annotation and labeling, visual search engines, robotic manipulation systems that understand verbal commands, visual question answering pipelines, content moderation systems, accessibility tools that describe image contents, and custom quality inspection systems that can be configured with natural language descriptions of defects rather than extensive training data.
Подробная версия доступна на английском языке.
Hailuo MiniMax
Hailuo MiniMax is a high-quality video generation model developed by the Chinese AI company MiniMax, distinguished by its impressive motion quality and ability to generate visually compelling video content with natural, fluid movement dynamics. Released in September 2024, Hailuo gained international recognition for producing some of the most realistic motion patterns among AI video models, particularly excelling in human movement, facial expressions, and complex physical interactions. The model supports both text-to-video and image-to-video modes, accepting natural language descriptions and reference images to create short clips with consistent visual quality and temporal coherence. Hailuo's transformer-based architecture processes multimodal inputs to generate content demonstrating strong understanding of physical world dynamics, including gravity, momentum, fabric movement, and environmental interactions. The model handles diverse content from photorealistic scenes to stylized artistic content, with particular strength in cinematic quality footage with professional-grade lighting and composition. Hailuo supports various output resolutions and aspect ratios suitable for social media, advertising, and creative projects across different platforms. The model demonstrates competitive performance in international benchmarks, often ranking alongside or above Western competitors in motion quality. As a proprietary model, Hailuo is accessible through MiniMax's platform and through fal.ai and Replicate, enabling integration into custom applications and production workflows. The model represents the growing strength of Chinese AI research in generative video technology.
Подробная версия доступна на английском языке.
Hunyuan Video
Hunyuan Video is a large-scale text-to-video AI model developed by Tencent with 13 billion parameters, making it one of the largest open-source video generation models available. Built on a Dual-stream Diffusion Transformer architecture that processes text and visual tokens through parallel attention streams before merging them, Hunyuan Video achieves exceptional visual quality with rich detail, accurate color reproduction, and strong temporal consistency across frames. The model supports both text-to-video generation from natural language descriptions and image-to-video generation where a static image is animated with contextually appropriate motion. Hunyuan Video produces videos at up to 720p resolution with smooth motion and physically plausible dynamics, generating content that stands out for its cinematic quality and aesthetic sophistication. The dual-stream architecture enables deep cross-modal understanding between text semantics and visual generation, resulting in strong prompt adherence for complex scene descriptions involving multiple objects, spatial relationships, and specific motion patterns. The model handles diverse content types including realistic scenes, animated styles, abstract visualizations, and nature footage with consistent quality. Released under the Tencent Hunyuan License which permits both research and commercial use with certain conditions, the model is available on Hugging Face and supported by the Diffusers library ecosystem. Key applications include professional video content creation, advertising and marketing video production, social media content generation, visual concept prototyping for film and animation studios, and educational content creation. Hunyuan Video particularly excels at generating aesthetically pleasing compositions with attention to lighting, depth of field, and cinematographic principles.
Подробная версия доступна на английском языке.
Hunyuan-DiT
Hunyuan-DiT is a bilingual text-to-image diffusion transformer model developed by Tencent, featuring a Diffusion Transformer architecture designed for high-quality image generation with native Chinese and English language understanding. The model employs a transformer-based diffusion approach that replaces the traditional U-Net backbone used in earlier diffusion models with a more scalable and efficient transformer architecture. Hunyuan-DiT uses a bilingual CLIP text encoder combined with a multilingual T5 encoder to process prompts in both Chinese and English with deep semantic understanding. The model generates high-resolution images with strong compositional accuracy, detailed textures, and faithful prompt adherence across various artistic styles including photorealism, traditional Chinese painting, modern illustration, and digital art. Its training dataset includes extensive Chinese cultural content, enabling it to accurately render Chinese characters, traditional artistic motifs, architectural elements, and cultural scenes that most Western-trained models cannot handle properly. Hunyuan-DiT supports controllable generation through various conditioning mechanisms and can produce images at multiple resolutions and aspect ratios. Released as open-source under a permissive license, the model is available on Hugging Face and GitHub with full training and inference code. It requires GPUs with 11GB or more VRAM for efficient operation. Chinese technology companies, digital content creators in Chinese-speaking markets, researchers in multilingual AI, and artists exploring cross-cultural visual creation form its primary user base. Hunyuan-DiT represents Tencent's significant contribution to the open-source image generation ecosystem and advances the state of bilingual visual AI.
Подробная версия доступна на английском языке.
I2VGen-XL
I2VGen-XL is a high-quality image-to-video generation model developed by Alibaba DAMO Academy that produces video content with strong semantic and temporal coherence from single input images. Released in November 2023, I2VGen-XL employs a cascaded architecture decomposing video generation into two stages: a base stage generating low-resolution video with correct semantic content and motion patterns, followed by a refinement stage that upscales and enhances visual quality for the final output. This two-stage approach lets the model first focus on understanding content and motion dynamics before applying detailed visual refinement, resulting in videos maintaining both semantic accuracy and visual quality. The model demonstrates strong capabilities in preserving the identity and visual characteristics of the input image while generating plausible temporal evolution, making it effective where maintaining visual consistency with source material is critical. I2VGen-XL handles diverse input types including photographs of people, animals, landscapes, and artistic compositions, applying contextually appropriate motion patterns respecting physical properties and spatial relationships in the original image. The model generates videos with smooth frame transitions, consistent lighting, and natural motion dynamics avoiding artifacts common in earlier approaches. Key use cases include animated product showcases, dynamic content from stock photography, animating concept art and design mockups, and social media content with engaging visual motion. Available under the Apache 2.0 license, I2VGen-XL is accessible on Hugging Face and Replicate, offering a capable open-source solution for image-to-video generation that balances quality with computational efficiency.
Подробная версия доступна на английском языке.
IC-Light
IC-Light (Intrinsic Compositing Light) is an AI relighting model developed by Lvmin Zhang, the creator of ControlNet, that manipulates and transforms lighting conditions in photographs with remarkable realism. Built on a Stable Diffusion backbone with specialized lighting conditioning, the model with over one billion parameters can take any photograph of an object or person and completely alter the light source direction, color temperature, intensity, and ambient lighting while maintaining photorealistic shadows, highlights, and surface reflections. IC-Light operates in two distinct modes: foreground relighting where the subject is extracted and relit independently, and background-compatible relighting where the lighting is adjusted to match a new background environment. The model understands physical light behavior including specular reflections, subsurface scattering on skin, metallic surfaces, and transparent materials, producing results that respect real-world optical properties. IC-Light accepts text descriptions or reference images to define the target lighting setup, offering intuitive control over the final appearance. Released under the Apache 2.0 license, the model is fully open source and has been integrated into ComfyUI with dedicated workflow nodes. Professional photographers, product photographers, digital artists, and e-commerce teams use IC-Light for correcting unfavorable lighting in existing photos, creating studio-quality lighting from casual snapshots, matching product lighting across catalog images, generating dramatic cinematic lighting for creative projects, and preparing composited images with consistent illumination across elements.
Подробная версия доступна на английском языке.
160 моделей найдено · Sayfa 2 / 7