AI Dungeon

AI Dungeon is a text-based adventure and AI-guided roleplay platform from Latitude. Instead of following a finite script, the player supplies an action, spoken dialogue, or narrative instruction and a neural language model generates what happens next. The player can choose Do mode for physical actions, Say mode for character dialogue, or Story mode for direct narrative control. The system can introduce characters, locations, conflicts, discoveries, and consequences in response to the player’s direction, creating an open-ended story that can continue for as long as the user wishes. Players may start with a quick adventure, select from community-published scenarios, or create original scenarios with custom premises, characters, and settings. Adventures and scenarios can be saved, edited, revisited, shared, or played with others through multiplayer features. Creative controls include undo, redo, retry, and edit, allowing players to correct weak generations or steer the narrative toward a preferred outcome. Plot Components organize persistent guidance such as AI Instructions, Plot Essentials, Author’s Notes, and Story Summaries. Story Cards preserve important information about characters, places, objects, and lore, while the Memory System helps retain relevant details from earlier events. AI Dungeon also supports AI-generated images through its See mode and image features, with credits available on paid plans. Subscription tiers primarily increase access to models, context length, memories, image credits, and generation speed. The service is available online and through mobile applications, with an official Guidebook, Discord community, social channels, and scripting documentation for advanced scenario creators. The comparison workflow is especially valuable when a single prompt can be sent to several models and the decision depends on output quality rather than brand reputation. Users can use the available categories to narrow evaluation to the kind of artifact they actually need, compare alternatives visually, and return to leaderboards when a new model or benchmark result becomes relevant. This makes the site both a practical discovery surface and an ongoing reference for generative AI evaluation. The comparison workflow is especially valuable when a single prompt can be sent to several models and the decision depends on output quality rather than brand reputation. Users can use the available categories to narrow evaluation to the kind of artifact they actually need, compare alternatives visually, and return to leaderboards when a new model or benchmark result becomes relevant. This makes the site both a practical discovery surface and an ongoing reference for generative AI evaluation.