About our Basic AI Models

About our Basic AI Models

About our Basic AI Models

AI Dungeon relies on AI language models to drive game experiences. We also offer several AI image models to generate pictures during gameplay. Each of these models has been developed independently and has different characteristics.

In AI Dungeon, Basic models are available to all players. These currently include Griffin, MythoMax, Tiefighter, and Stable Diffusion 1.5. Basic text models are always unlimited and free.

Free players can spend Scales to speed-boost these text models for 1-hour or 24-hours.

Griffin

Griffin uses a 6 billion parameter GPT-J model developed by EleutherAI. It tends to be one of our more unexpected models, which has both positive and negative impacts on the experience. In some instances, Griffin might require more retries or editing. In others, it can add interesting, unexpected turns to your adventures and games.

MythoMax

MythoMax is a brand new Basic model specifically fine-tuned for roleplaying. Based on Llama 2 13B, it's twice as large as our current Griffin model, doubling the AI power that free users get unlimited access to. AI Dungeon is one of the few places free users can get unlimited access to AI storytelling, and we're excited to make that experience even better.

Tiefighter

Tiefighter is another brand new, open-source Llama 2 13B variant. It’s even more capable of expansive and creative roleplaying in a wide range of subjects. Alpha testers have described Tiefighter as a combination of Mixtral and MythoMax for how it focuses on plot versus character emotions. Tiefighter has been released as another unlimited Basic model, giving free players even more AI options.

Stable Diffusion 1.5

Stable Diffusion 1.5 was the first AI image model we added to AI Dungeon. It is capable of creating unique pictures from a text prompt. These generations cost 1 Credit for Wanderers (free players) and Travelers. Subscribers have unlimited access to this model.

icon
image

© Latitude 2023