Gaming AI apps

AI-powered text adventure that turns your every action into an open-ended story.

AI Dungeon is an AI-driven text adventure where the model acts as a dungeon master, generating an open-ended story that responds to whatever you type — any action, any genre, with no fixed script. Players can pick from preset settings or author custom scenarios and worlds, and the game keeps track of characters and events as the narrative unfolds across web and mobile.

No-code engine for AI characters that talk, remember and stay on-script.

Charisma.ai is a no-code platform for building interactive stories and conversational AI characters with voice, memory, emotion and goals. A controllable story-graph keeps characters on the writer's narrative while still allowing generative improvisation, and SDKs drop the characters into Unity, Unreal and the web. It's used for games, training simulations, brand experiences and digital companions.

AI agent for Unreal Engine — generate C++, Blueprints and 3D from prompts.

An AI toolkit that embeds directly inside Unreal Engine 5 to accelerate game development. From natural-language prompts it generates and explains C++ code, builds and analyzes Blueprints, produces textured 3D models, and answers context-aware questions about Unreal concepts. An agent mode can chain these into multi-step changes across a project.

AI-native RPG platform where every world is built and played from a prompt.

Voyage lets anyone describe a world in natural language and drops them inside it as a playable, text-based RPG. Built on Latitude's World Engine, characters carry their own motivations and memories, storylines emerge from player choices, and the world keeps evolving even after you log off. Up to four players can share a world in co-op.

AI agents that playtest your game and file the bugs.

modl.ai is an AI QA automation platform that tests games the way a player would — entirely through the screen, with no SDK or code integration. You describe tests in plain language ("complete the tutorial") and its agents explore the build, surfacing crashes, performance issues, and gameplay-logic bugs, then generate reports with visual evidence and severity scores. It also offers on-demand bots that simulate real players at varied skill levels for multiplayer balancing and load.

Real-time world model that generates interactive, explorable AI video.

Odyssey is a world-model startup whose AI generates interactive video you can steer in real time — explorable, 3D-like scenes streamed frame by frame without a game engine. Its current model, Odyssey-2, turns a text prompt or image into a multi-minute simulation you navigate as it is generated. It is positioned for film, gaming, and interactive media, and is available as a free research-preview experience.

Multi-agent AI platform that builds production 3D asset pipelines for games.

Atlas (Atlas AI Studio) orchestrates fleets of specialized AI agents that generate, texture, optimize, and engine-integrate 3D, 2D, and motion assets from natural-language descriptions and moodboards. Rather than producing isolated outputs from a single model, it assembles end-to-end production pipelines aimed at studio workflows, with exports to Unreal and Blender. The Vienna-based company targets games, CGI, and immersive worldbuilding.

Turn 2D images into production-ready 3D models.

An AI 3D-asset production platform that converts 2D images and concept art into game-ready 3D models with clean topology. Pairs generative AI with a human-in-the-loop QA step so output meets production specs, custom art styles, and pipeline requirements. Used across games, e-commerce, product design, and AR/VR by studios including Aardman, Rebellion, and The Sandbox.

AI assistant that builds and dresses 3D game environments.

An AI assistant for professional game artists and environment designers that reasons over a studio's own asset library to generate, populate, and dress complex 3D scenes. It runs inside existing editors via plugins for Unreal Engine, Unity, 3ds Max, Maya, and Blender — and works without uploading assets to the cloud. Rather than generating images, it automates the repetitive placement and worldbuilding work behind large productions.

Create and play AI-generated games from a prompt.

Astrocade is an AI-native platform for creating and playing games without code. Describe an idea and its AI generates playable experiences — handling art, animation, music, sound, and gameplay logic — then publishes them to a social feed where others can discover and play. Founded by brothers Amir and Ali Sadeghian, it reached over 20 million users within eight months of launch.

AI roleplay games set in your favorite story worlds.

Hidden Door is a social AI storytelling platform where players roleplay collaborative adventures guided by an AI Narrator, presented as interactive graphic novels. It launched with public-domain worlds like Pride and Prejudice, The Call of Cthulhu, and The Wizard of Oz, and licenses IP through revenue-share with creators rather than training on their work without permission. Every playthrough differs because choices reshape the story.

A generative behavior engine for living game characters.

Artificial Agency's behavior engine uses foundation models to turn game systems into autonomous agents at runtime — characters that remember player interactions and personalize encounters, plus 'game director' systems that adapt story and pacing. Studios integrate it into any genre without game-specific training, keeping their existing tools and workflows. The engine launched in alpha to select studios via a pilot program.

Conversational AI characters and NPCs for games and virtual worlds.

Convai is a developer platform for building lifelike, voice-driven non-player characters and virtual humans. Its SDKs for Unity, Unreal Engine, PlayCanvas, and the web add end-to-end speech, LLM reasoning, long-term memory, perception, and action execution so characters can hold natural conversations, understand their surroundings, and perform in-world tasks. Studios use it to ship interactive NPCs across games, XR, and the open web.

AI-native game engine and platform for building games with generative AI.

Series AI is a generative-AI game development company building Rho Engine, an AI-native, full-stack platform that uses generative models to speed up game creation — from concept and assets through to playable builds. Founded by veteran game developer Pany Haritatos, it works with studios and IP holders to co-develop AI-native games.

Real-time generative world and video models (Oasis, Lucy).

Decart is an AI lab building real-time generative world and video models, including Oasis (a fully AI-generated, playable Minecraft-style open world) and Lucy (real-time video transformation used in gaming, ads, and virtual try-on). Its Oasis 3 is positioned as an API-accessible world model for Physical AI, with a low-latency inference stack. Products are accessed via the platform API and interactive demos.

AI production platform for game creative, ads, and live-ops assets.

Layer is an AI production platform that helps game studios scale creative output — user-acquisition ads, live-ops content, and marketing assets — across image, video, 3D, and audio, with custom style training for on-brand results. It supports batch generation of 500+ variants per day and integrates with Photoshop, Blender, Maya, Unreal, Unity, and Figma. It is used by 300+ entertainment brands including Zynga, King, and SciPlay.

Build playable 3D browser games with natural-language vibe coding.

Rosebud AI lets anyone build playable browser games — 2D, 3D, and multiplayer — by describing them in natural language ('vibe coding'), with no installs or manual coding required. It bundles a Game Creator plus tools like PixelVibe for AI game assets, an RPG/Visual Novel maker, and AI Storybook. Over 500,000 games have reportedly been created on the platform.

AI game ideation, market research, and asset generation.

Ludo.ai generates game concepts, surfaces market and trend research, and helps developers validate ideas. It has since broadened into asset generation (sprites and spritesheets, 2D/3D models, audio, and video) alongside its ideation and market-analysis features. Access is via a web app, REST API, an MCP integration, and a Unity Editor plugin.

Train custom models to generate style-consistent game assets.

Scenario lets game developers train custom models on their own art references (5-100 images) to generate style-consistent assets across image, video, 3D, and audio. It offers a node-based workflow editor, 550+ underlying models, and an API plus MCP server for agent integration. It is used in production by studios including Ubisoft, Unity, and Scopely.