13 categories and 89 products and companies, charted by an ai agent.
Text-to-video generation produces video content from text prompts or descriptions. Filmmakers, marketers, content creators, and studios use it to generate or prototype video sequences.
runway gen-4.5 · veo 3.1 · seedance 2.5 · kling 3.0 · wan 2.7 · minimax h3 · luma ray3 · pika 2.2 · ltx-2.5 pro · hailuo minimax
Open-source video foundation models are large neural networks trained on video data that developers can download and modify to build custom applications. Video researchers, ML engineers, and companies building video tools use these models as starting points for tasks like understanding video content or generating frames.
wan2.2 · cogvideox · open-sora · animatediff · ltx-video
Text-to-image generation creates visual content from natural language descriptions using AI models. Used by designers, marketers, content creators, and enterprises for rapid asset creation.
flux 3 · midjourney v7 · dall-e 4 · stable diffusion 3.5 · ideogram 3 · adobe firefly 4 · imagen 4 · recraft v4
Image-to-video animation generates video sequences from static images, with motion and transitions applied. Used by content creators, marketers, and studios to produce animated videos from photographs or artwork.
kling i2v · runway act-one · luma ray3 modify · pika pikaffects · wan i2v
runway video-to-video · adobe premiere ai · topaz video ai · descript · capcut ai
heygen · synthesia · did · runway act-one avatars · pika lip sync · d-id · colossyan · hedra · captions
invideo ai · creatify · arcads · lumen5
adobe firefly inpaint · clipdrop · magnific ai · krea ai · genmo
Foundation video models are large neural networks trained on video data that can be fine-tuned or used as backbones for downstream video tasks; researchers and engineers use them to build video generation, understanding, and editing applications.
sora · veo · kling · wan alibaba · seedance · luma ray · runway · pika · genmo mochi · lightricks ltx · luma ai · seedance bytedance
Foundation image models are neural networks trained on large image datasets that generate, edit, or understand images; used by developers, designers, and creative applications to build image generation and manipulation features.
midjourney · flux black forest labs · stable diffusion · imagen · ideogram · recraft · firefly · leonardo · dall-e gpt image · flux
Apps that generate videos from images or image sequences, applying motion, transitions, and temporal coherence to static visual content. Used by content creators, marketers, and video editors to produce dynamic video outputs from still images.
capcut · higgsfield · openart · krea · freepik · adobe firefly · canva
Open model hosting platforms that run video and image generation models on user-submitted prompts, returning generated media files. Used by developers, researchers, and end users who want to run generative models without local infrastructure.
fal · replicate · comfyui · civitai · hugging face
3D world generation creates three-dimensional environments and scenes from text, images, or other inputs for use in games, films, architecture visualization, and virtual environments. Game developers, film studios, architects, and metaverse platforms use these tools to build explorable 3D spaces.
world labs · meshy · tripo · skybox ai