Videos

Models 109

• Gen 3
  Qwen 3.5 122B A10B

  By Alibaba

  Qwen3.5-122B-A10B is a larger Mixture-of-Experts Qwen3.5 model with 122B total parameters and 10B active per token, targeting higher peak capability while staying compute-efficient at inference.

  📷Images🎥Videos📚Stories
  NewMultimodal

  Released 1mo ago
• Gen 3
  Qwen 3.5 Flash

  By Alibaba

  Qwen3.5-Flash is the speed and cost optimized Qwen3.5 variant designed for high-throughput chat and multimodal prompting with a very long context window.

  💬Chatting📷Images🎥Videos
  NewMultimodal

  Released 1mo ago
• Gen 3Seedance
  Seedance 2.0

  By ByteDance

  Seedance 2.0 is ByteDance's multimodal AI video model that turns text plus image, video and audio references into high-resolution, sound-synced clips, giving creators director-level control over camera, motion, style and multi-shot storytelling.

  🎥Videos🖼️Image generation📚Summaries
  NewMultimodal

  Released 1mo ago
• Gen 2
  Ming Flash Omni 2.0

  By InclusionAI

  Ming-flash-omni 2.0 is an open sparse MoE omni-modal model that unifies text, image, video and audio understanding and generation, using a Ling-2.0 Mixture-of-Experts backbone with 100B parameters and about 6B active per token.

  🖼️Image generation🎵Music📷Images🎥Videos
  NewMultimodal

  Released 1mo ago
• Gen 4
  GameGen X

  By GameGen-X

  GameGen-X is a diffusion transformer specifically built for open-world game video, generating and interactively controlling characters, environments, and actions in long gameplay clips.

  🎥Videos📄Writing legal documents🕹️Adventure games
  NewVideo

  Released 1mo ago
• Gen 4
  XMAX X1

  By Xmax AI

  XMAX X1 is a real-time interactive video model that fuses virtual and real worlds, using the phone camera and touch gestures for millisecond-level, on-device AR-style experiences.

  🎥Videos📢Prompts📚Summaries📚Novels
  NewVideo

  Released 1mo ago
• Gen 3Kling
  Kling 3.0

  By Kuaishou Technology

  Kling Video 3.0 is Kuaishou's newest AI video model that unifies text, image, audio and reference video in one engine, generating up to 15 second photorealistic clips with native multi-language audio and strong consistency across shots.

  🎥Videos📷Images📚Stories
  NewMultimodal

  Released 1mo ago
• Gen 3LongCat
  LongCat Flash Omni

  By Meituan

  560B parameter omni-modal MoE model (about 27B active) for real time audio-visual interaction, built on LongCat-Flash with multimodal perception and speech modules.

  🏋️Fitness📷Images💬Chatting🎥Videos
  NewMultimodal

  Released 1mo ago
• Gen 3
  MOVA

  By OpenMOSS

  Open source foundation model that jointly generates video and audio in one pass, achieving tightly synchronized lip movements and environment-aware sound effects.

  🎥Videos🔊Text to speech🎵Music🎬Animations
  NewVideo

  Released 1mo ago
• Gen 3
  Vidu Q3

  By ShengShu Technology

  Vidu Q3 is ShengShu Technology’s long-form AI video model that generates a single 16-second clip with native, synchronized audio and 1080p video in one generation.

  🎥Videos📽️Image to video🎤Lip sync videos
  NewMultimodal

  Released 1mo ago
• Gen 3
  SkyReels V3

  By Skywork AI

  Long-form video extension engine that analyzes scene semantics and motion to extend clips with coherent shots, maintaining strong temporal consistency and cinematic storytelling

  🎥Videos🎬Video editing📚Stories📝Content
  NewMultimodal

  Released 2mo ago
• Gen 4Grok
  Grok Imagine 1

  By xAI

  Grok Imagine is xAI’s video-audio generative model, exposed through the Imagine API, that turns text or images into cinematic videos and supports text-to-image, text-to-video, image-to-video and rich video editing with strong instruction following, latency and cost performance.

  🎥Videos📢Prompts📷Images👗Fashion
  NewVideo

  Released 2mo ago
• Gen 3Kimi
  Kimi K2.5

  By Moonshot AI

  Kimi K2.5 is Moonshot’s native multimodal open model, trained on about 15T vision-text tokens to power advanced coding and visual reasoning and to orchestrate self-directed agent swarms with up to 100 sub-agents and 1,500 tool calls.

  💻Coding📝Writing📷Images🎥Videos
  NewMultimodal

  Released 2mo ago
• Gen 4PixVerse
  PixVerse v5.6

  By Pixverse

  PixVerse V5.6 is PixVerse’s latest video model, upgrading V5.5 with cinema level visuals, more natural multilingual voices, smoother physics aware motion and less warping, while keeping generation speed and cost roughly the same as earlier V5 models.

  🎥Videos📷Images🔧Code optimization
  NewVideo

  Released 2mo ago
• Gen 7Gemma
  MedGemma 1.5

  By Google

  MedGemma 1.5 4B is Google’s updated 4B-parameter medical vision-language model that improves CT, MRI and histopathology understanding while remaining compute efficient for offline and cloud healthcare text and imaging workflows.

  🌍Internationalization🎥Videos📚Stories
  NewText

  Released 2mo ago
• Gen 4LTX
  LTX 2

  By LTX Studio

  LTX-2 is Lightricks’ 19B diffusion-based audio-video foundation model that generates synchronized 4K video and stereo audio from text or images, with distilled and LoRA variants for faster local generation.

  🎥Videos🔍SEO content🎵Music📽️Image to video
  NewVideo

  Released 2mo ago
• Gen 3Seed
  Seed 1.8

  By ByteDance

  General-purpose agentic model that unifies LLM and VLM abilities with search, code execution and GUI control, designed to perceive text and images, reason over long contexts and autonomously carry out multi-step tasks.

  🤖Task automation💻Coding📷Images🎥Videos
  Multimodal

  Released 3mo ago
• Gen 7Gemma
  T5Gemma 2

  By Google

  T5Gemma 2 is Google’s next generation Gemma 3 based encoder-decoder family, a lightweight multilingual and multimodal LLM that reads text and images, outputs text, and offers 128K context with tied embeddings and merged attention for efficient on-device deployment.

  📷Images💻Coding📝Writing🎥Videos
  Text

  Released 3mo ago
• Gen 3Wan
  Wan 2.6

  By Alibaba

  Wan 2.6 is Alibaba’s latest Wan AI multimodal video model, turning text, images, audio and short reference clips into up to 15 s 1080p videos with native audio sync, multi-shot storytelling, and strong character and style consistency.

  🎥Videos🖼️Image generation📷Images
  Multimodal

  Released 3mo ago
• Gen 4LongCat
  LongCat Video Avatar

  By MeiGen AI

  LongCat-Video-Avatar is Meituan’s audio-driven avatar model built on LongCat-Video, generating super-realistic, lip-synced long videos from audio plus optional text and images, with stable identity, natural motion and support for multi-person scenes.

  🎥Video avatars🎥Videos
  Video

  Released 3mo ago
• Gen 4
  Molmo 2

  By Ai2

  Molmo 2 is AI2’s open 4B/7B/8B multimodal model for images and video, delivering state-of-the-art grounded video QA, pointing and tracking that return coordinates and timestamps for events instead of text-only answers.

  🎥Videos📝Writing📷Images🔍Multimodal search
  Text

  Released 3mo ago
• Gen 4Runway Gen
  Runway GWM 1

  By Runway AI

  Runway GWM-1 is a family of general world models built on Gen-4.5 that generate action-conditioned video in real time, powering explorable simulated environments, conversational avatars and robotics simulators for training and interactive applications.

  🎥Videos💬Conversational avatars
  Video

  Released 3mo ago
• Gen 4Kling
  Kling 2.6

  By Kuaishou Technology

  Kling Video 2.6 is Kling AI's latest video model that natively generates video plus dialogue, music and sound effects in one step, turning text or images into 5-10 second 1080p clips with tightly synced audio-visual storytelling for creators and advertisers.

  🎥Videos🔊Text to speech🎵Music
  Video

  Released 3mo ago
• Gen 7Mistral
  Ministral 3 14B

  By Mistral AI

  The largest Ministral 3 model offers frontier text and vision capabilities comparable to larger 24B models. Edge-optimized for single GPU deployment (24GB VRAM in FP8), it delivers state-of-the-art performance for chat, document analysis, and complex reasoning tasks with multilingual support across 40+ languages.

  💬Chatting📋Document chat🎥Videos
  Text

  Released 3mo ago
• Gen 7Mistral
  Ministral 3 8B

  By Mistral AI

  Best-in-class text and vision model for edge deployment, optimized for single GPU operation with minimal footprint. Features interleaved sliding-window attention for efficient inference. Ideal for constrained environments, chat interfaces, image/document understanding, and balanced local deployment scenarios.

  📷Images📝Writing💬Chatting🎥Videos
  Text

  Released 3mo ago
• Gen 7Mistral
  Ministral 3 3B

  By Mistral AI

  The smallest yet robust Ministral model, edge-optimized for ultra-low-resource environments. Despite its compact size (~3GB), it provides strong language and vision capabilities, outperforming older 7B models. Runs entirely in browser via WebGPU. Ideal for IoT devices, mobile apps, and offline assistants.

  📷Images👤Avatars📢Prompts🎥Videos
  Text

  Released 3mo ago
• Gen 4PixVerse
  PixVerse V5.5

  By Pixverse

  PixVerse V5.5 is PixVerse’s audio-visual text and image to video model that generates 5-10 s 1080p multi-shot clips with native speech, music and SFX, improved motion stability and multi-shot camera control for story driven, lip-synced short videos.

  🎥Videos🔊Text to speech🎥Short videos
  Video

  Released 3mo ago
• Gen 7Mistral
  Mistral Large 3

  By Mistral AI

  Mistral 3 is Mistral AI’s next-gen open multimodal, multilingual family, combining small dense Ministral 3 edge models with the frontier Mistral Large 3 MoE to deliver image-aware, long-context language intelligence.

  📷Images💻Coding📝Writing🎥Videos
  Text

  Released 3mo ago
• Gen 4Runway Gen
  Gen 4.5

  By Runway AI

  Runway Gen-4.5 is Runway's latest text-to-video and image-to-video model, ranked #1 on independent benchmarks for motion, realism and prompt adherence, delivering cinematic, physics-aware clips with fine camera, style and timing control for creators and studios.

  🎥Videos📽️Image to video
  Video

  Released 3mo ago
• Gen 4Kling
  Kling o1

  By Kuaishou Technology

  Kling Video O1 is Kling AI’s unified multimodal video model that fuses text-to-video generation with image- and video-based editing, using advanced reasoning and motion control to create short, high-quality cinematic clips in a single workflow.

  🎥Videos🎬Video editing📽️Image to video
  Video

  Released 3mo ago
• Gen 4
  Vidi2

  By ByteDance

  Vidi2 is ByteDance’s second generation large multimodal video model for understanding and creation, adding fine grained spatio temporal grounding, long video retrieval, and video question answering so it can find both the right time ranges and object boxes from natural language queries.

  🎥Videos🎥Video summaries🤣Memes
  Video

  Released 4mo ago
• Gen 4Pika
  Pika 2.5

  By Mellis

  Pika 2.5 is an upgraded text and image to video model that delivers sharper detail, smoother motion, better lip-sync and physics, plus more precise control over camera, pacing, and aspect ratios for social and production clips.

  🎥Videos📷Images🔄Workflows
  Video

  Released 4mo ago
• Gen 4Hunyuan
  HunyuanVideo-1.5

  By Tencent

  HunyuanVideo-1.5 is Tencent's 8.3B-parameter open-source video diffusion model for text-to-video and image-to-video generation, delivering high-quality, stable motion clips while running efficiently on consumer-grade GPUs.

  🎥Videos💻Coding📷Images
  Video

  Released 4mo ago
• Gen 4SAM
  SAM 3

  By Meta Platforms

  SAM 3 is Meta’s third-generation Segment Anything foundation model that performs promptable segmentation and tracking in images and videos, finding all instances of open-vocabulary concepts from text or visual prompts.

  🖼️Image generation💻Coding📷Images🎥Videos
  Image

  Released 4mo ago
• Gen 4
  Dia 2

  By Nari Labs

  Dia2 is an open source streaming dialogue TTS model that generates speech in real time from partial text, supports audio conditioning for natural back and forth conversations, and ships 1B and 2B checkpoints under Apache 2.0.

  🔊Text to speech💬Chatting🎥Videos🗒Transcription
  Audio

  Released 4mo ago
• Gen 7
  Hermes 4.3

  By Nous Research

  Hermes 4.3 is Nous Research’s 36B hybrid reasoning model, based on Seed-OSS-36B, offering long context (up to 512k) and very high helpfulness on RefusalBench while staying locally deployable.

  📝Writing💬Chatting🎥Videos💼Personal assistant
  Text

  Released 4mo ago
• Gen 7ERNIE
  ERNIE 4.5 VL 28B A3B Thinking

  By Baidu

  A multimodal MoE model that “looks, reads, and reasons” across images, video, and text. It adds tool use and a Thinking with Images mode, supports long context, and activates about 3B parameters per token for flagship-level VLM quality at practical latency.

  📷Images💻Coding👤Avatars🎥Videos
  Text

  Released 4mo ago
• Gen 4LongCat
  LongCat-Video

  By Meituan

  LongCat-Video is a long-horizon text and image to video model that keeps identity, style, and motion consistent over extended clips. It supports in-place edits, camera and pacing control, and fast previews to delivery-quality renders.

  🎥Videos🎬Video dubbing🎬Video editing
  Video

  Released 5mo ago
• Gen 4Vidu
  Vidu Q2 Turbo

  By ShengShu Technology

  Speed tuned Q2 variant for rapid drafts and cost efficient iteration.

  🎥Videos
  Video

  Released 5mo ago
• Gen 4Vidu
  Vidu Q2 Pro

  By ShengShu Technology

  Premium Q2 profile with maximum temporal stability and micro detail.

  🎥Videos🎬Video editing🎥Video summaries
  Video

  Released 5mo ago
• Gen 4Vidu
  Vidu Q2

  By ShengShu Technology

  Vidu Q2 is an upgraded text and image to video model. It delivers sharper detail, steadier motion, stronger identity and style locking, and precise control over camera and pacing. It supports text to video, image to video, and in-place edits for fast iteration to production-ready clips.

  🎥Videos🎬Video editing📽️Image to video🎥Video summaries
  Video

  Released 5mo ago
• Gen 4Nova
  Nova Reel

  By Amazon

  Nova Reel is the video model for text or image to video and in place edits. It delivers stable motion, identity and style locking, and precise control of camera and pacing for clips that drop into real production timelines.

  🎥Videos🎬Video editing📽️Image to video
  Video

  Released 5mo ago
• Gen 4Veo
  Veo 3.1 Fast

  By Google

  Veo 3.1 Fast is the speed tuned variant of Veo 3.1. It trades a little peak fidelity for much lower latency and cost, keeping the same controls for camera, motion, and style so teams can iterate rapidly.

  🎥Videos✈️Travel📚Stories🔄Workflows
  Video

  Released 5mo ago
• Gen 4Veo
  Veo 3.1

  By Google

  Veo 3.1 is a high fidelity text and image to video model. It delivers sharper detail, steadier motion, stronger identity and style locking, and precise control of camera and pacing for production ready clips.

  🎥Videos📽️Image to video
  Video

  Released 5mo ago
• Gen 3Grok
  Grok Imagine v0.9

  By xAI

  Ling-flash-2.0 is a high-speed multilingual instruction model built for very low latency and high throughput. It supports long context, tool and function calling, and clean JSON outputs, which makes it ideal for live chat, voice assistants, and real-time automation.

  📷Images📝Writing🔄Paraphrasing🎥Videos
  Multimodal

  Released 5mo ago
• Gen 4
  VGAN

  By VGAN

  VGAN, the 2016 Video GAN, generates short clips directly from noise using a spatio temporal GAN. It learns a static background and a moving foreground with a learned mask, then a 3D conv discriminator judges realism in space and time.

  🎥Videos
  Video

  Released 5mo ago
• Gen 4Sora
  Sora 2 pro

  By sora AI

  Flagship text to video profile for complex scenes, realistic physics, and long, stable takes.

  🎥Videos🎬Video editing📽️Image to video
  Video

  Released 5mo ago
• Gen 3Sora
  Sora 2

  By OpenAI

  Sora 2 is OpenAI’s next-generation text/image-to-video model. It produces sharper, longer, more stable clips with better physics, identity and style locking, and precise control over camera and motion—built for fast iteration and production delivery.

  🎥Videos🔍SEO content📚Stories
  Video

  Released 5mo ago
• Gen 4Hailuo
  Hailuo 2.3

  By MiniMax

  Hailuo 2.3 is a high-quality text-to-video model for longer, steadier shots and stronger physics

  🎥Videos💻Coding📷Images
  Image

  Released 6mo ago
• Gen 4Wan
  Wan 2.5

  By Alibaba

  Wan 2.5 is the next-gen text-to-video system. It delivers sharper detail, longer and more stable shots, stronger physics, and tighter identity/style locking, with precise control over camera and motion. It supports text→video, image→video, and in-place edits for fast iteration to production-ready clips.

  🎥Videos📷Images🎥Presentations💼Sales
  Video

  Released 6mo ago