Generative models

A generative probabilistic model that represents each document as a finite mixture over topics, fitted with variational approximate inference and EM, and evaluated on document modelling, text classification and collaborative filtering.

A 2006 paper by Hinton, Osindero and Teh that uses complementary priors and a greedy layer-by-layer algorithm to train deep belief nets, then fine-tunes them; the resulting three-hidden-layer model both modelled and classified handwritten digits well.

A generative adversarial network pits a generator that makes fake data against a discriminator that judges it, both trained in a zero-sum game; proposed by Ian Goodfellow and colleagues in 2014, it led to a series of architectures for generating images.

Goodfellow, Bengio, and co-authors proposed a generative-model framework with adversarially trained generator and discriminator.

This 2018 paper introduces a style-based GAN generator that separates high-level attributes from stochastic variation and allows scale-specific control, improving distribution quality, interpolation, and disentanglement.

A 2020 paper by Mildenhall, Srinivasan, Tancik, Barron, Ramamoorthi and Ng that represents a scene as a neural radiance field: a network mapping 5D coordinates to density and colour, optimized from a sparse set of images with known camera poses.

A 2021 paper by Rombach, Blattmann, Lorenz, Esser and Ommer that runs diffusion models inside the latent space of pretrained autoencoders, cutting the compute needed for high-resolution image synthesis while adding cross-attention for text and bounding-box control.

An open-weight text-to-image model released publicly by Stability AI on 22 August 2022, built on latent diffusion and small enough to run on consumer GPUs, which made local image generation widely accessible.

AlphaFold 3 is a structure-prediction model from Google DeepMind and Isomorphic Labs, announced on 8 May 2024. It predicts the structures and interactions of proteins, DNA, RNA, ligands and chemical modifications, extending structure prediction beyond proteins alone.[^1]

ESM3 is a protein generative model announced by EvolutionaryScale in June 2024 that reasons over sequence, structure and function together, and generated a fluorescent protein estimated to sit over 500 million years of evolution away from known ones.

π0 is a generalist robot policy from Physical Intelligence, built by adapting a 3-billion-parameter vision-language model with flow matching so one model can output fast, continuous motor commands for dexterous manipulation.

ByteDance's video generation model released in February 2026, built to produce picture and sound together from text, image, audio or video inputs, and known for a copyright dispute that paused its global rollout.

Unsupervised learning trains algorithms on data that carry no labels, letting them find structure such as clusters, low-dimensional representations or generative models. It underlies clustering tools and neural generative models from the Hopfield network and Boltzmann machine to the variational autoencoder.

A Boltzmann machine is a neural network whose configurations are scored by a global energy and trained by a stochastic learning algorithm. Introduced by Geoffrey Hinton and Terry Sejnowski, it brought energy-based models into cognitive science and machine learning.

A 1985 Cognitive Science paper by Ackley, Hinton and Sejnowski that presents a general learning rule for connectionist networks and reports simple examples of efficient internal representations.

Mila, the Quebec AI Institute, is a Montreal deep-learning institute founded in 1993 by Yoshua Bengio as LISA; its graduates include Ian Goodfellow and Hugo Larochelle.

Topic models represent documents as mixtures of latent topics, giving a compact statistical description of text collections. The 2003 Latent Dirichlet Allocation paper formalized this idea as a hierarchical Bayesian model.

Deep neural network acoustic modeling uses multilayer networks to map speech to units such as phones. A 2010 NIPS paper used Deep Belief Nets and a mean-covariance RBM to extract speech features, reaching 20.5% phone error on TIMIT.

A 2010 NIPS paper by Dahl, Ranzato, Mohamed and Hinton applies Deep Belief Nets and a mean-covariance RBM to acoustic modeling, reaching 20.5% phone error on speaker-independent TIMIT phone recognition.

Kuaishou Technology is a Beijing-based short-video and live-streaming company, founded in 2011 as GIF Kuaishou. It listed in Hong Kong in 2021 and builds the Kling video generation model.

A variational autoencoder is a generative model whose encoder maps an input to a distribution over a latent space and whose decoder reconstructs the input from a sample; training maximises a lower bound on the data likelihood.

A 2013 paper by Diederik P. Kingma and Max Welling that introduced a stochastic variational inference and learning algorithm for directed latent-variable models with intractable posteriors, making approximate inference practical on large datasets.

Insilico Medicine is a drug discovery company founded in the United States in 2014 that pairs an AI software platform, Pharma.AI, with its own pipeline of drug programmes and automated laboratory work.

A DeepMind generative model that predicts raw audio one sample at a time, bringing neural text-to-speech close to human quality and later powering Google Assistant and Cloud Text-to-Speech voices.

WaveNet is a deep neural network that generates raw audio waveforms one sample at a time, predicting each sample from all the ones before it. Used for text-to-speech, it was rated by listeners as more natural than the systems it was compared with.

Deepfake is synthetic media: images, video or audio made or altered with AI. The term, a blend of "deep learning" and "fake", appeared in late 2017 and now names a whole class of face-swapping and manipulation techniques.

CycleGAN learns image-to-image translation between two domains without aligned pairs, combining an adversarial loss with a cycle-consistency constraint to constrain the mapping.

A 2017 paper by Karras, Aila, Laine and Lehtinen that trains generative adversarial networks by growing the generator and discriminator from low to high resolution, improving stability, quality and variation.

A 2017 paper by van den Oord, Vinyals and Kavukcuoglu proposing VQ-VAE, a generative model whose latent code is discrete rather than continuous, and whose prior is learned rather than fixed, so that useful representations can be learned without supervision.

Runway is a New York-based AI video company founded in 2018. It co-released Stable Diffusion in 2022, has built a series of video models from Gen-1 to Gen-4.5, and in February 2026 raised $315 million at a $5.3 billion valuation.

A 2018 paper by Andrew Brock, Jeff Donahue and Karen Simonyan that trained generative adversarial networks at the largest scale attempted at the time, introduced BigGANs and set a new state of the art in class-conditional image synthesis.

Stability AI is a British generative-AI company founded in 2019 in London. It is known for releasing the open-weight Stable Diffusion image model in 2022, and it has since faced copyright lawsuits over training data.

A 2020 paper by Jonathan Ho, Ajay Jain and Pieter Abbeel that trained a U-Net to reverse a gradual noising process, making diffusion models a practical rival to GANs and paving the way for DALL-E 2, Imagen and Stable Diffusion.

Diffusion models learn to reverse a gradual noising process, so they can turn random noise into a new image by removing noise step by step. Latent diffusion, which runs this in a compressed space, became the basis of Stable Diffusion.

DALL·E is OpenAI's text-to-image model unveiled on 5 January 2021: a 12-billion-parameter transformer trained on 250 million text-image pairs that generates a picture token by token from a caption.

A 2021 OpenAI paper describing a single 12-billion-parameter autoregressive transformer trained on 250 million image-text pairs, which generates images from text and, evaluated zero-shot, is competitive with domain-specific models.

MiniMax is a Shanghai-based AI company founded in December 2021 by computer-vision researchers from SenseTime. It builds text, speech and video models and consumer apps such as Talkie and Hailuo AI, and listed in Hong Kong in January 2026.

Midjourney is a San Francisco-based independent research lab that runs an image generator. Founded and led by David Holz, it has never taken outside funding and has been involved in several copyright lawsuits.

DALL·E 2 is an image generator that OpenAI announced on 6 April 2022. It uses a diffusion decoder conditioned on CLIP image embeddings and supports image variations and language-guided editing. OpenAI deprecated its API in late 2025 and shut it down in May 2026.

Imagen is a text-to-image diffusion model announced by Google Research in May 2022, known for pairing a frozen large language model with diffusion and for the DrawBench benchmark; Google did not release it publicly.

Imagen is a text-to-image diffusion model that uses a large pretrained language model to encode prompts; it reached a zero-shot FID of 7.27 on COCO and introduced the DrawBench benchmark.

Midjourney is an AI image generation service from Midjourney, Inc. Its open beta began on 12 July 2022 through a Discord bot, and it later added a website, video generation and new model versions.

A 2022 paper by William Peebles and Saining Xie that replaces the U-Net backbone of diffusion models with a transformer operating on latent patches, showing that image quality improves predictably as compute grows and reaching a state-of-the-art FID of 2.27 on ImageNet 256×256.

PixVerse, founded in 2023 as Aishi Technology, is an AI video generation company whose in-house models power a web app, an API and agent tools; a March 2026 Series C made it a unicorn.

Runway Gen-2 is a multimodal video generation system announced in March 2023 that turns text, images or video clips into new video, and was among the first commercially available text-to-video models.

Midjourney V5 is a version of the Midjourney text-to-image model, released in alpha on 15 March 2023 and refined through the 5.x line before being succeeded by V6 in December 2023.

SDXL is Stability AI's open text-to-image diffusion model, announced in July 2023 with a 3.5B-parameter base model and a refiner stage, generating natively at 1024×1024. A distilled version, SDXL Turbo, cut generation down to a single step.

DALL·E 3 is OpenAI's image generation model announced on 20 September 2023 and released inside ChatGPT that October. It turned a user's idea into a detailed prompt before drawing, and was later replaced by GPT Image.

Black Forest Labs is a German AI lab based in Freiburg im Breisgau, founded in 2024 by Robin Rombach, Andreas Blattmann and Patrick Esser. It develops the FLUX image models and raised a $300 million Series B in December 2025.

Sora was OpenAI's text-to-video model, previewed in February 2024 and released that December, able to generate clips up to a minute long. It was succeeded by Sora 2 and discontinued in 2026.

Stable Diffusion 3, announced by Stability AI on 22 February 2024, is a suite of 800 million to 8 billion parameter models that pairs a diffusion transformer with flow matching.

Hailuo AI is MiniMax's generative AI product, launched in March 2024 for text and music generation and extended to video with the video-01 model. It drew copyright scrutiny when major studios sued MiniMax in September 2025.

Veo is a generative video model from Google DeepMind, announced at Google I/O on 14 May 2024. Google described it as its most capable video model, able to produce 1080p clips longer than a minute and later succeeded by Veo 2.

Kling (可灵) 1.0 is Kuaishou's video generation model. Text-to-video opened for testing on 6 June 2024, and Kling could generate videos of up to two minutes at up to 1080p.

Kling AI is Kuaishou's AI video and image creation platform, built on its Kling models. It began testing in June 2024 and went global that July, and its annualised revenue run rate passed US$240 million by December 2025.

Runway's 2024 video generation model, trained jointly on video and images, with finer temporal control such as keyframing; later extended with an API, video-to-video and Act-One performance capture.

FLUX.1 is a family of 12-billion-parameter text-to-image models released by Black Forest Labs on 1 August 2024 in pro, dev and schnell variants. The dev version is open-weight, and the variants carry different licenses.

Amazon Nova is Amazon's family of foundation models, introduced on 3 December 2024 and offered exclusively in Amazon Bedrock. It spans text understanding models from Micro to Pro, plus Canvas for images and Reel for video, and was succeeded by Nova 2 in December 2025.

HunyuanVideo is Tencent's open-source text-to-video generation model, with over 13 billion parameters, released in December 2024. Running it locally needs at least 60GB of GPU memory.

Step-Video-T2V is an open text-to-video model with 30 billion parameters. StepFun released its weights, code, a Turbo variant and a technical report under the MIT license on 17 February 2025.

Wan2.1 is Alibaba's open-source video generation model family, released on 25 February 2025 under Apache 2.0. It creates videos from text or images and can render Chinese and English text.

Runway Gen-4 is a video generation model released on 31 March 2025 that keeps characters, locations and objects consistent across shots, working from reference images and text instructions rather than per-project training.

Kling 2.0 is Kuaishou's AI video generator, launched globally in April 2025 and billed by the company as the world's most powerful, with a multimodal editing feature and the Multi-modal Visual Language concept.

π0.5 is a vision-language-action model from Physical Intelligence, released in April 2025, built to generalize to unfamiliar homes and trained by co-training on many kinds of robot and web data.

Veo 3 is Google's video generation model, announced at Google I/O on 20 May 2025, that generates video with native audio, including dialogue with lip syncing; its outputs carry SynthID watermarks.

FLUX.1 Kontext is a family of image models from Black Forest Labs, announced on 29 May 2025, that generates and edits pictures from both text and image inputs. Its open-weight dev variant followed in June 2025.

Seedance 1.0 is a video-generation model from ByteDance Seed, announced in June 2025. One model handles text-to-video and image-to-video, produces 1080p multi-shot clips, and ranked first on two Artificial Analysis leaderboards.

Genie 3 is Google DeepMind's general-purpose world model, announced on 5 August 2025. It turns text prompts into interactive 720p environments that can be navigated in real time at 24 frames per second, and was released as a limited research preview.

Nicknamed "nano-banana", Gemini 2.5 Flash Image is Google's image generation and editing model released on 26 August 2025, able to blend images, keep characters consistent across edits and apply targeted natural-language edits.

Seedream 4.0 is ByteDance Seed's image model released in September 2025, which puts text-to-image generation, image editing and multi-image composition into a single architecture and can output images from 1K up to 4K resolution.

An open-source image generation model from Tencent that handles image understanding and generation inside a single autoregressive mixture-of-experts framework; its code, weights and technical report were released in September 2025.

Sora 2 is OpenAI's second-generation video and audio generation model, unveiled on 30 September 2025 with a social iOS app. OpenAI discontinued Sora in 2026, and its API models were scheduled to shut down in September 2026.

ERNIE 5.0 is Baidu's natively omni-modal foundation model, unveiled in November 2025 and described as having 2.4 trillion parameters. It is proprietary and was made available to enterprises and consumers in January 2026.

FLUX.2 is Black Forest Labs' image generation and editing model family, released on 25 November 2025, spanning hosted tiers and open-weight releases, with support for 4-megapixel output and up to 10 reference images.

Kling O1 is a video model that Kuaishou announced on 1 December 2025. The company described it as the first unified multimodal video model: text, image, video and subject inputs go into one engine used both for generation and for editing.

Kling 3.0 is Kuaishou's video and image model series, launched on 5 February 2026 as four models. It generates up to 15 seconds of video with simultaneous audio, adds multi-shot generation and supports native 4K output.

Gemini Omni Flash is a Google DeepMind generative model that makes and edits video from text, image, audio and video prompts. Video creation arrived in the Gemini app on 19 May 2026, and the model later became the recommended replacement for Google's Veo 3.1 API previews.

Vidu S1 is Shengshu's real-time interactive video generation model, released in July 2026. It takes spoken commands as control input, runs on consumer GPUs at 540p and up to 42 frames per second, and keeps generating without drift.

FLUX 3 is a multimodal model family from Black Forest Labs, announced on 23 July 2026, covering images, video, audio and action prediction. Its video model became generally available on 4 August 2026.

Vidu S2 is a two-part video generation system: S2-Avatar renders interactive digital characters at 720p in real time, and S2-Editing transforms live video streams. Both also produce stereo views for immersive displays.

PixVerse R2 is a real-time world model announced on 22 September 2026, an upgrade to R1 that keeps a navigable world generating while the user moves inside it and holds state across a session.

Kling 4.0 is a text-to-video generation model released to all users in October 2026, adding longer shots, multi-image references and more convincing physical motion, alongside a faster Flash variant built for high-volume work.

Suno is an AI music generator that turns prompts into finished songs, and increasingly a browser-based production suite with a studio, voice recording and stem separation.

ElevenLabs is an AI voice and media platform whose business spans voice generation, voice agents for customer experience and creative tools; it ended 2025 with $350 million in annual recurring revenue.

Ian Goodfellow is a machine learning researcher and first author of the 2014 paper that introduced generative adversarial networks, which train a generator and a discriminator against each other. He has worked at OpenAI, Google Brain, Apple and Google DeepMind.

Runway builds generative video, image and audio tools used by more than 60 million creative users, and researches general-purpose world models that simulate video, audio and interfaces in real time.

Durk Kingma is a machine learning researcher and co-author of the variational autoencoder and the Adam optimizer; he has worked at OpenAI, Google Brain / DeepMind and Anthropic.

Emad Mostaque is a British-Bangladeshi former hedge fund manager who founded Stability AI in 2019 and served as its CEO until he resigned in March 2024.

Cheng Yixiao is a Chinese entrepreneur who co-founded Kuaishou, served as its chief product officer, and has been its chief executive officer since 2021 and chairman since 2023.

Cristóbal Valenzuela is a Chilean-born technologist and the co-founder and CEO of Runway, the New York company behind the Gen series of AI video models.

A 2018 paper by Diederik P. Kingma and Prafulla Dhariwal presenting Glow, a flow-based generative model built on invertible 1x1 convolutions, which improved log-likelihood on standard benchmarks and could synthesise and manipulate large images.

Luma AI is a generative AI company founded in 2021 in Redwood City, California, known for the Dream Machine video app, the Ray3 video model and the Luma Agents platform, with more than $1.1 billion raised.

Midjourney V1 was Midjourney's first image-generation model, released in February 2022 and used through a bot on the company's Discord server. It produced abstract, painterly images and was replaced by V2 and V3 within months.

Midjourney V3, released on 25 July 2022, was the third model version of Midjourney; it raised image quality sharply by learning from how users behaved, and it produced the image that won a Colorado State Fair digital-art prize.

Midjourney V4 is the fourth-generation image generation model released in alpha on 5 November 2022. It was the first Midjourney model trained on Google TPUs, and reviewers described it as a large step up from V3.

Stable Diffusion 2 is an open-source text-to-image model family released by Stability AI on 24 November 2022, built on a new OpenCLIP text encoder and shipped with upscaling, depth-guided and inpainting models; version 2.1 followed weeks later.

Text watermarking embeds a hidden statistical signal in generated text so that a short span of tokens can later be identified as machine output, while the text still reads normally to people.

Runway Gen-1 is a video-to-video diffusion model released in February 2023 that restyles existing footage from text or image prompts while keeping the source video's structure. It was among the first commercially available video-generation models.

Stable Audio, launched by Stability AI on 13 September 2023, is a text-to-music and sound generator that Stability describes as its first audio product, later joined by open and enterprise versions.

v0 is Vercel's tool that turns a written description of an interface into frontend code built on React, Tailwind CSS and shadcn/ui. It drew 100,000 waitlist sign-ups and moved from Alpha to Beta in October 2023.

Lyria is Google DeepMind's family of music-generation models, first announced in November 2023. It writes instrumental and vocal music from prompts, carries inaudible SynthID watermarks, and has reached users through YouTube Shorts, Vertex AI and the Gemini API.

Stable Video Diffusion is Stability AI's image-to-video model, released on 21 November 2023 as a research preview that turns a single still image into a clip of a few seconds.

Google's second-generation text-to-image model, released in December 2023, whose standout feature was rendering legible text and logos inside generated images.

Midjourney V6 is a version of Midjourney's image generation model, released in alpha on 21 December 2023 and trained from scratch over nine months; its 6.1 update followed in July 2024.

A Google DeepMind generative world model with 11 billion parameters that turns text, images or sketches into controllable virtual worlds, described as the first generative interactive environment trained on unlabelled internet videos without action labels.

Hunyuan-DiT is Tencent's open text-to-image diffusion transformer, built for fine-grained English and Chinese prompt understanding. Its 2024 report and weights were released publicly, and human evaluations placed it ahead of several open models but behind leading closed ones.

Google's text-to-image model announced at I/O in May 2024, which Google described as its highest-quality image generator, with fewer visual artifacts and improved text rendering.

Dream Machine is Luma AI's text- and image-to-video generator, released to the public on 12 June 2024. It later became the home of Luma's Ray3 and Ray3.14 video models.

Kolors is Kuaishou's open-weight text-to-image model, released on 6 July 2024, built on latent diffusion and tuned for both Chinese and English prompts.

CogVideoX is an open-source family of text-to-video models from Zhipu AI and Tsinghua University, released starting in August 2024 and derived from the earlier open-source CogVideo.

Kling 1.5 is a September 2024 version of Kuaishou's video generation model, adding 1080p output, a Motion Brush, and quality improvements before being succeeded by Kling 1.6 in December 2024.

A closed text-to-image model from Black Forest Labs, released on 2 October 2024 and available only through an API. Black Forest Labs said it generates six times faster than its predecessor, and it reached the top of the Artificial Analysis image arena leaderboard.

Movie Gen is a set of media foundation models Meta premiered on 4 October 2024. It covers video generation, personalised video, precise video editing and audio generation, and was not released publicly.

A family of text-to-image models from Stability AI, announced in October 2024: the 8.1-billion-parameter Large, a 4-step Turbo variant and the 2.5-billion-parameter Medium, released under a community licence.

Kling 1.6 is a December 2024 upgrade to Kuaishou's Kling video generation model, with better handling of motion and camera instructions in prompts and improved visual quality; it was succeeded by Kling 2.0 in April 2025.

Genie 2 is a foundation world model from Google DeepMind, announced on 4 December 2024, that turns a single prompt image into a playable 3D environment that a person or an AI agent can explore with keyboard and mouse controls.

Veo 2 is a video-generation model Google announced on 16 December 2024, emphasising real-world physics and human movement, with outputs up to 4K that can extend to minutes.

NVIDIA Cosmos is a platform of world foundation models that generate physics-based video to train robots and autonomous vehicles, released as open-weight models in January 2025.

Hunyuan3D 2.0 is Tencent's large-scale system for generating high-resolution textured 3D assets: a shape diffusion transformer plus a texture model, released with code and pretrained models on 21 January 2025 alongside its technical report.

Step-Audio is an open family of speech-interaction models released by StepFun on 17 February 2025. It includes Step-Audio-Chat, with 130 billion parameters, and Step-Audio-TTS-3B.

Midjourney V7 is an image generation model released as an alpha to Midjourney's community on 4 April 2025, the first of the company's models to have personalization on by default and the first to add a cheap, fast Draft Mode.

A speech-to-speech foundation model in Amazon's Nova family, announced on 8 April 2025, which handles recognition, understanding and speech generation in a single model rather than a chain of separate ones.

Seedream 3.0 is ByteDance Seed's Chinese–English image-generation foundation model of 2025, improving on Seedream 2.0 in prompt alignment, Chinese text rendering and native 2K output. Seedream 4.0 succeeded it in September 2025.

Kling 2.1 is a Kuaishou video generation model released at the end of May 2025 in three tiers — 720p Standard, 1080p High Quality and 1080p Master — with claimed gains in motion, physical simulation and prompt responsiveness. It was superseded by Kling 2.5 Turbo.[^1][^2][^3]

Imagen 4 is a Google text-to-image model announced on 20 May 2025 that outputs images at up to 2K resolution with improved fine detail and typography, shipping in the Gemini app, Whisk, Vertex AI and Workspace before its API models were retired in 2026.

Midjourney's first video model, released on 18 June 2025, turns an existing image into a short clip of about five seconds. It was presented as the company's first step toward real-time open-world simulation.

Runway Aleph is a video model released in July 2025 that edits an input video by adding, removing or transforming objects, generating a new angle of a scene, or changing style and lighting, across many visual tasks.

HunyuanWorld 1.0 is Tencent's open-source model for generating explorable, mesh-exportable 3D worlds from text or images. Released in July 2025, it was followed by a series of lighter and faster versions.

An open family of video generation models released in July 2025 under the Apache 2.0 licence, notable for bringing a two-expert mixture-of-experts design to video diffusion and for a 5B model that runs on a single consumer GPU.

A 20-billion-parameter open-weights image generation model released by Qwen in August 2025, notable for rendering text inside images, especially Chinese.

MAI-Voice-1 is Microsoft AI's speech generation model, announced in August 2025 as the first model in the MAI family and built to produce expressive, high-fidelity audio quickly.

Kling 2.5 Turbo is a Kuaishou video generation model released at the end of September 2025, which Artificial Analysis ranked No. 1 worldwide for text-to-video and image-to-video ten days after launch, and which cut creators' per-video generation cost by nearly 30%.

Ray3 is a video generation model released by Luma AI on 18 September 2025 and described by the company as the first reasoning video model, with HDR output, native 1080p and a neural upscaler to 4K.

MAI-Image-1 is Microsoft's first image generation model developed entirely in-house, announced on 13 October 2025 and rolled out in Bing Image Creator and Copilot from 4 November 2025.

Veo 3.1 is an October 2025 update to Google's Veo 3 video model, adding audio across Flow's features plus new editing tools; its API previews were retired in October 2026.

Nano Banana Pro (Gemini 3 Pro Image) is Google's image generation and editing model built on Gemini 3 Pro, announced on 20 November 2025. It renders legible text, pulls live information from Search, and outputs up to 4K.

Runway Gen-4.5 is a video generation model released in December 2025 that topped the Artificial Analysis text-to-video leaderboard and later gained native audio, multi-shot editing and a role as the base for Runway's GWM-1 world model.

Kling 2.6 is a Kuaishou video generation model released in December 2025 that produces visuals, voiceover, sound effects and ambient audio in a single pass, with clips capped at 10 seconds.

GWM-1 is Runway's first general world model: an autoregressive system that generates 720p video frame by frame in real time and can be steered with camera pose, robot commands or audio.

An audio-video generation model from Volcano Engine, announced on 18 December 2025, that produces picture, voices, music and ambient sound together and keeps them in millisecond-level sync.

Nano Banana 2 is Google's product name for Gemini 3.1 Flash Image, an image generation and editing model in the Gemini API, released in preview on 26 February 2026 and later replaced by Nano Banana 2.1.

Midjourney V8, released as an alpha on 17 March 2026, generates images about five times faster than earlier versions, adds a native 2K mode, and improves rendering of text inside images.

MAI-Image-2 is Microsoft AI's second-generation text-to-image model, released to developers on 2 April 2026 in Microsoft Foundry and the MAI Playground, with faster generation and stronger lighting, skin tones and in-image text.

Stability AI's 2026 audio generation family: three open-weight models and a hosted flagship, able to generate and edit music and sound effects up to about six minutes long.

Muse Image and Muse Video are Meta Superintelligence Labs' first media generation models, announced on 7 July 2026. Muse Image launched immediately, and Muse Video entered early preview.

Seedance 2.5, released by ByteDance in July 2026, is an audio-video joint generation model that makes clips of up to 30 seconds per pass, accepts many multimodal references, and supports editing parts of a scene without regenerating the whole clip.

Anastasis Germanidis is a co-founder of Runway, the generative AI video company he started in 2018, and a co-author of the research behind Gen-1, Runway's video-to-video generation model.

Patrick Esser is a researcher on generative image and video models: a co-author of the latent diffusion paper that became the basis of Stable Diffusion, lead author of the Stable Diffusion 3 paper, and a 2024 co-founder of Black Forest Labs.

Andreas Blattmann is a researcher on generative image models, second author of the latent diffusion paper behind Stable Diffusion, co-author of the Stable Diffusion 3 paper, and a 2024 co-founder of Black Forest Labs, which released the FLUX.1 models.

First author of the latent diffusion paper that underlies Stable Diffusion, and later co-founder and CEO of Black Forest Labs, the company behind the FLUX image models.

Pika is a set of AI creative tools for video, image and audio, reached through a web app as well as an iOS app, a Model Context Protocol server and a ChatGPT app, and it lists its own models alongside third-party ones.

Synthesia is an AI video platform for business: users type a script and an AI avatar presents it, with tools for generation, dubbing, translation, screen recording and motion graphics, plus a separate Sessions line for interactive roleplay and surveys.

HeyGen is an AI video platform that turns a short recording of a person into a speaking avatar for social content, courses and business videos. Its Avatar V model is pitched at consistency across long clips and many languages.

David Holz is the founder of Midjourney, a San Francisco independent research lab whose text-to-image tool became one of the best-known AI art generators. He earlier co-founded the hand-tracking hardware company Leap Motion.

Midjourney V2 was the second model version of Midjourney's image generation model, released on 12 April 2022 and the default model until V3 arrived that July. Midjourney later described it as colorful and painterly but low in coherency.

Voxtral TTS is Mistral AI's first text-to-speech model, released in March 2026: about 4B parameters, nine languages, cloning from a three-second clip, open weights for non-commercial use and an API priced per character.