Contents

Appendix · The Ecosystem

Appendix · The Ecosystem

The whole map, one company at a time

The book covers the model labs in depth. This directory zooms out to the entire creative stack forming around them: the tools for writing, producing, finishing, serving, and distributing AI-native work. Each entry gets a one-line brief, and the ones the book profiles link straight to their chapter.

250 companies · 5 phases · kept current by the update pipeline

Development

Everything before a frame is generated: writing, structure, and pre-production.

Story Development

  • Anthropic's LLM; strong for long-form story development, structure, and drafting. (This guide was written with it.)

    ~$30B+ raised

  • OpenAI's assistant, widely used for ideation, outlines, and script drafts.

    Part of OpenAI

  • Google's assistant, tied into the Gemini media models.

    Part of Google

  • Fiction-focused writing tool with story-aware drafting features.

    ~$3M raised

  • Writing assistant for grammar, tone, and clarity.

    ~$1.5B raised

  • Readability editor that flags dense or passive prose.

  • AI code editor; used by technical teams building creative pipelines.

    ~$3.5B raised

  • AI screenwriting and script-analysis tool.

  • Screenwriting software with collaboration and AI assists.

  • AI-assisted writing and development tool.

Storyboarding

  • End-to-end AI storyboarding and previs from a script.

    ~$335M (Lightricks)

  • Generates storyboards from a script or prompt.

  • AI storyboarding and pre-production workspace.

  • Turns scripts into storyboards and animatics.

  • Storyboarding tool with AI frame generation.

    ~$100K raised

  • AI storyboarding and shot-planning tool.

Production Assistant

  • AI script breakdown and scheduling for production.

    ~$2M raised

  • Production management: breakdowns, call sheets, scheduling.

  • Production planning and collaboration suite.

  • AI virtual-production backgrounds and 2.5D environments.

    ~$400K raised

  • AI script-to-storyboard and previs assistant.

  • AI production-planning assistant.

  • Screenwriting and pre-production toolkit.

  • AI script analysis, breakdown, and scheduling.

    ~$1.4M raised

  • AI production and creative assistant.

    ~$100K raised

  • AI-assisted production tooling.

Production

Where pixels and sound are generated. The core of the guide.

Image Models

  • Frontier text-to-image with a distinctive, highly aesthetic house style.

    Bootstrapped, no outside funding

  • OpenAI's reasoning image model; plans and self-checks before rendering. Top of the image arena.

    Part of OpenAI

  • Google's Gemini-based image line that replaced Imagen; tiers from Lite to 4K Pro.

    Part of Google

  • Open-weight-friendly frontier image models; FLUX.2 is the current base.

    ~$450M raised

  • Strong typography and text rendering; 4.0 shipped open weights with structured prompting.

    ~$100M raised

  • Fast-rising frontier image lab; Reve 2.1 ranks near the top of the arena.

    ~$390M raised

  • xAI's image and video generation inside Grok.

    Part of xAI

  • Maker of Stable Diffusion; the open-weights foundation of the ecosystem.

    ~$230M raised

  • Adobe's commercially-safe generative models, embedded across Creative Cloud.

    Part of Adobe

  • Realtime, creator-focused generation; shipped its own Krea 2 foundation model.

    ~$83M raised

  • Design-oriented image model with vector and brand-style control.

    ~$42M raised

  • Alibaba's image model line; strong text and layout.

    Part of Alibaba

  • ByteDance's image model with precise editing and layer output.

    Part of ByteDance

  • MIT-licensed pixel-native open image model.

    ~$70M raised

  • Microsoft's in-house image model; MAI-Image-2.5 ranks top-tier.

    Part of Microsoft

Video Models

  • The model that launched the video era; consumer app retired, API sunsets Sept 2026.

    Part of OpenAI

  • Google's video line, folded into the multimodal Gemini Omni; tops the video arenas.

    Part of Google

  • Frontier Chinese video model; strong motion, native audio, ~$18B valuation.

    Part of Kuaishou

  • Pioneer video lab; Gen-series models, the Aleph editing model, and a full suite.

    ~$860M raised

  • Physics-strong, cost-efficient video model.

    Public (HKEX)

  • Alibaba's open-weight video family, widely used in ComfyUI.

    Part of Alibaba

  • ByteDance's video model; 2.5 does native single-pass 30-second 4K.

    Part of ByteDance

  • Dream Machine and Ray models plus a physical-AI research effort.

    ~$1.1B raised

  • xAI's video model with native audio; 1.5 briefly led image-to-video.

    Part of xAI

  • HappyHorse (Alibaba ATH)

    Tops the open video arenas with single-pass joint audio and video.

    Part of Alibaba

  • Tencent's open video/image family.

    Part of Tencent

  • Consumer-friendly video app with effects and social focus.

    ~$135M raised

  • Fast consumer video generator, strong on effects.

    ~$500M raised

  • Chinese video model with reference-to-video and audio.

    ~$400M raised

  • Open-weight, real-time-leaning video model with audio.

    ~$335M raised

  • Cinematic video model (Marey); team joined Reka for physical AI.

    ~$155M raised

  • Open-source video model from the GLM/Zhipu lineage.

    Part of Zhipu AI

  • Real-time and world-model video (Mirage, Oasis).

    ~$450M raised

  • Emerging video generation startup.

  • Maker of the open Mochi video model.

    ~$30M raised

  • Character motion and controllable video.

    ~$19M raised

  • Creator-focused video generation and Superstudio.

    ~$10M raised

Realtime Video

  • Live, latency-optimized generation as you type or paint.

    ~$83M raised

  • Sub-40ms/frame streaming diffusion for interactive video.

    ~$450M raised

  • Decentralized video infrastructure with realtime AI.

    ~$52M raised

  • Realtime interactive generation research.

Voice & Dubbing

  • Leading AI voice synthesis and dubbing; expanding into music and SFX.

    ~$820M raised

  • Realtime voice synthesis and voice agents.

    Part of Meta

  • Voice and AI-NPC platform for games and interactive media.

    ~$130M raised

  • AI dubbing and localization for film and TV.

    ~$26M raised

  • AI voiceover for narration and corporate content.

    ~$12M raised

  • High-fidelity voice cloning used in film and games.

    ~$4M raised

  • Emotionally-expressive voice interface (EVI).

    ~$73M raised

  • Low-latency state-space voice models (Sonic).

    ~$190M raised

  • Open voice/speech research lab (Moshi).

    ~$330M (nonprofit lab)

  • Open-leaning TTS and voice cloning.

  • Natural conversational voice model.

    ~$390M raised

  • Emotive AI voice acting for content.

    ~$40M raised

  • AI dubbing for video localization.

    ~$0.8M raised

  • Creative asset platform with AI voice and music.

    ~$48M raised

  • Google's research tool; popularized AI audio overviews.

    Part of Google

Lip Sync & Avatar

  • AI avatar video for talking-head and marketing content.

    ~$75M raised

  • Enterprise avatar video with multilingual narration.

    ~$540M raised

  • Expressive character video and talking avatars.

    ~$45M raised

  • Lip-sync API to match any voice to any face.

    ~$7M raised

  • Mobile-first AI video, captions, and avatars.

    ~$100M raised

  • Talking-photo and avatar video platform.

    ~$50M raised

  • Also ships lip-sync and character effects.

    ~$135M raised

  • Text-to-speech, expanding into avatars.

  • Avatar video for training and L&D.

    ~$30M raised

  • Voice and avatar content platform.

    ~$12M raised

  • Browser video editor with avatars and AI tools.

    ~$35M raised

  • Realtime talking-avatar video.

    ~$11M raised

  • Expressive avatar and character video.

    ~$100M raised

  • AI talking-avatar content tool.

Aggregators & Interfaces

  • Unified canvas across many image/video models plus its own.

    ~$83M raised

  • Multi-model image playground and community.

    ~$35M raised

  • Creator platform with fine-tunes; part of Canva.

    Part of Canva

  • Upscaling and enhancement, plus a multi-model canvas; part of Freepik.

    Part of Freepik

  • Fast-growing multi-model video/image platform with cinematic controls.

    ~$140M raised

  • Aggregator for video and image generation.

    ~$3M raised

  • Multi-model generation suite.

  • Game-asset generation with custom trained models.

    ~$12M raised

  • Doc-style AI audio/video editor (also in editing).

    ~$100M raised

  • Prompt-to-video for social and marketing.

    ~$50M raised

  • Creator video and multi-tool Superstudio.

    ~$10M raised

  • Multi-model AI video and image tools.

  • Multi-model creative platform.

  • AI short-film and drama generation suite.

    Part of Kunlun Tech

Infinite Canvas

  • The node-based workflow tool that became the power-user standard.

    ~$47M raised

  • Node/canvas workspace chaining image, video, and text models.

    ~$52M raised

  • Node-based creative canvas for pro pipelines.

    ~$4M raised

  • Canvas-first image workspace for designers.

    ~$3M raised

  • Infinite-canvas generative workspace.

    ~$2M raised

  • Canvas plus best-in-class upscaling.

    Part of Freepik

  • Multi-model canvas and aggregator.

Multi-Shot & Narrative

  • Character-consistent multi-shot storyboarding and video.

    ~$2M raised

  • Narrative

    Multi-shot AI storytelling workflow.

  • ByteDance's multi-shot content generator.

    Part of ByteDance

  • Long-form multi-shot story generation.

  • Multi-shot narrative video.

  • Flick / FLIK

    Multi-shot AI film tools.

    ~$6M raised

  • Composable AI mini-apps and workflows.

    ~$17M raised

  • AI studio for narrative production.

  • Story planning and multi-shot generation.

  • AI narrative and animation studio tooling.

  • Multi-shot generation research.

    ~$3.5M raised

  • Multi-shot product and ad video.

    ~$14M raised

Music Generation

  • Frontier text-to-song music generation.

    ~$1.2B raised

  • Frontier music generation with high fidelity.

    ~$70M raised

  • Stability's music model; long-form generation.

    Part of Stability AI

  • Music generation from the voice leader.

    Part of ElevenLabs

  • Composition-focused music AI for scoring.

    ~$2.5M raised

  • Open-leaning music generation.

    Acquired by Google

  • Royalty-free adaptive music for video.

    ~$2.4M raised

  • Generative music streams and API.

    ~$4M raised

  • Customizable royalty-free AI music.

    ~$5M raised

  • Prompt-to-song music model.

    ~$0.5M raised

  • AI music creation platform.

    Part of Google

  • AI music generation and library.

Sound Effects

  • Text-to-sound-effects generation.

    Part of ElevenLabs

  • Video-to-audio foley generation (open research).

    Open research project

  • AI vocals and sound production.

    ~$20M raised

  • AI sound-effect generation.

    ~$41M raised

  • Adobe's generative audio tools.

    Part of Adobe

3D Production

  • Text/image-to-3D asset generation, popular with game and AR teams.

    ~$450M raised

  • Fast image-to-3D mesh generation.

    ~$400M raised

  • Luma's text-to-3D.

    Part of Luma AI

  • Open 3D generation from Tencent.

    Part of Tencent

  • High-quality 3D asset generation.

  • 2D-to-3D production models for games.

    ~$15M raised

  • Web-native 3D design with AI.

    ~$25M raised

  • Parametric real-time 3D asset generation.

    ~$3M raised

  • Generative 3D creation suite.

    ~$3.5M raised

  • AI-assisted virtual world and set building.

  • Character and animation pipeline tools.

    Public (TPEx: 6882)

  • Photoreal VFX and virtual production on phone.

    ~$1M raised

  • Industry-standard 3D DCC, adding AI assists.

    Part of Autodesk

World Models

  • Realtime, interactive, persistent generated 3D worlds.

    Part of Google

  • Fei-Fei Li's spatial-intelligence lab; Marble launched commercially.

    ~$1.2B raised

  • World-model foundation for robotics and simulation.

    Part of NVIDIA

  • Meta's world-generation research.

    Part of Meta

  • Realtime playable world models (Oasis).

  • Interactive world/video models; well funded.

    ~$340M raised

  • Also pursuing general world models (GWM).

    ~$860M raised

  • Spatial/world-model startup.

    ~$13M raised

Agentic Chat & Programmatic

  • Conversational interface to generate and edit media.

    Part of Luma AI

  • Agentic chat for media creation.

  • Programmatic video rendered from React code.

    ~$500K raised

  • API for templated, data-driven video.

  • Cloud video-editing API for automation.

    ~$750K raised

  • Generate video from a JSON spec via API.

  • Video understanding and generation API.

AI Ads & Social

  • AI UGC-style ads with avatars, at scale.

    ~$16M raised

  • Automated performance-ad generation.

  • Brand-safe enterprise content generation.

    ~$205M raised

  • Marketing copy and content platform.

    ~$130M raised

  • AI ad-creative generation.

  • AI ads and store content for SMBs.

  • AI creators and character-driven social video.

    ~$500K raised

Interactive

  • Interactive AI characters for chat and roleplay.

    ~$193M raised

  • AI TV-episode generation (Fable).

  • Story-to-illustrated-narrative engine.

  • Interactive AI narrative worlds.

  • AI character social platform.

    ~$8M raised

  • Interactive AI storytelling.

  • Interactive AI character experiences.

    ~$15M raised

Post-Production

Finishing: upscaling, grading, VFX, editing, and rights.

Upscaling

  • Long-standing image/video upscaling and enhancement.

    Bootstrapped

  • Creative upscaler that adds plausible detail.

    Part of Freepik

  • Fast creative upscaling in the Krea suite.

    Part of Krea

Color Grading

  • AI color grading for film and video.

  • Cloud color grading with AI looks.

  • Automatic photo/video colorization.

  • AI photo editing and color for photographers.

    ~$34M raised

Gen VFX

  • Photoreal face/de-aging VFX; used in major films.

    ~$25M raised

  • Generative VFX pipeline in Autodesk Flow.

    Part of Autodesk

  • High-end deepfake VFX studio.

    ~$20M raised

  • Vanity-AI and VFX for episodic TV.

    ~$5M raised

  • AI relighting and compositing (SwitchLight).

    ~$5M raised

  • Markerless motion capture from video.

    ~$13M raised

  • Gen VFX tools: inpainting, motion brush, editing.

    ~$860M raised

AI Editing

  • AI-native editing surface.

    Part of Runway

  • Dominant consumer editor with deep AI features.

    Part of ByteDance

  • Edit video/audio by editing the transcript.

    ~$100M raised

  • Auto-clips long video into shorts.

    ~$50M raised

  • Auto-captions and short-form editing.

    Bootstrapped

  • Mass-market design with generative video/editing.

    ~$600M raised

  • Long-to-short repurposing.

  • AI visual dubbing and dialogue editing (TrueSync).

    ~$32M raised

  • AI-assisted brand video editor.

    ~$20M raised

  • AI assistant editor for rough cuts.

  • Consumer editor with AI tools.

    Public (SZSE: 300624)

Legal & Likeness

  • Likeness protection and takedown for public figures.

  • Deepfake detection for enterprises.

  • Invisible watermarking and provenance.

  • Certification for models trained on licensed data.

  • Rights and content-provenance tooling.

  • Likeness and IP protection at scale.

Inference

The compute layer that serves every model.

Inference & Hosting

  • Fast inference for generative media; developer favorite.

    ~$335M raised

  • Run/fine-tune open models via API; acquired by Cloudflare.

    ~$58M raised

  • Production inference infrastructure and autoscaling.

    ~$285M raised

  • Inference and training cloud for open models.

    ~$1.3B raised

  • Low-cost, fast image inference API.

    ~$66M raised

  • The hub for open models, datasets, and inference.

    ~$400M raised

  • Cheap, fast image-generation API.

    ~$16M raised

  • GPU cloud for inference and training.

    ~$93M raised

  • Data and model versioning for ML teams.

    ~$7M raised

  • GPU marketplace and inference.

    ~$4M raised

Distribution

Where AI-native content reaches audiences.

Streamers & Platforms

  • Primary home for AI-made content, adding native gen tools.

    Part of Google

  • Studio incumbent experimenting with generative production.

    Public (NASDAQ)

  • AI-native streaming platform for short cinematic content.

    ~$2M raised

  • Platform for AI-native films and creators.

    ~$4M raised

  • Escape

    AI content streaming/community.

  • AI-native short-form platform.

Microdramas

  • AI-generated vertical microdrama platform.

  • AI microdrama studio and app.

    ~$150M (SpoonLabs)

  • Vertical short-drama streaming.

    Public (NYSE American)

  • Short-drama app leaning into AI production.

    ~$13M raised

  • My Muse

    AI microdrama and companion content.

  • Short-form vertical drama.

    ~$12M raised

  • Shortly

    AI-assisted microdrama platform.

  • mochi

    Short-form AI video app.

  • Market-leading English-language vertical microdrama app; original productions for Western audiences on coin-unlock monetization.

    Part of Crazy Maple Studio

  • One of the two global leaders and the largest catalog; a rare internationally profitable operator at scale.

  • Fast-growing coin-monetized microdrama app, reported as the fastest revenue riser in 2026.

  • Global vertical short-drama app competing in the top tier of the coin-monetized field.

  • DramaWave

    Vertical microdrama app rounding out the leading field behind ReelShort and DramaBox.

  • Fox-backed studio; My Drama uses AI for dubbing and subtitles on live-action, while its MyMuse hub produces fully-AI series.

    ~$22M raised

  • AI microdrama platform expanding across Southeast Asia with multi-language AI-generated series.