Enable Beta Testing Features: Enables access to features in development and not yet released. As a beta tester, you can experiment with features we plan to launch and review our latest improvements to AI Dungeon.

Note that these features are likely incomplete and may contain bugs. If you do find bugs, leave a report on our Feedback Site.

Click Here to learn more about Beta Features.

Safe Mode: The AI doesn’t understand what its outputs actually mean, so it could generate literally anything, including potentially obscene things.

Safe Mode moderates explicit content in both the user's input and the AI's output.

Note that turning this off does NOT allow you to break the Terms of Service.

Show Do / Say / Story Button: Show the button for Do / Say / Story, or hide it in favor of using /do, /say, /story commands instead. This may be useful if you only ever use one mode.

Auto Line Break in Story Mode: Toggle this setting on to add line breaks at the top of each Story input instead of continuing from the end of the previous line. This could be a helpful way to separate Story actions during gameplay and make it easier to read through your Adventure later.

You can also manually create line breaks by hitting shift-enter in the input box or hitting the return key on your mobile-phone keyboard.

Show Icon Text: Show text labels beneath the command buttons. Turning this off is an excellent way to free up screen space if you know what they are already.

Display Command Buttons: Allows you to toggle which command buttons are visible and what order they’re arranged in. Very useful for getting rid of buttons you don’t want to use.

Low Bandwidth Mode: Disables images. Toggle this on if you have slower internet or want faster performance.

Show Feedback Popup: Opt in to give us feedback on the AI periodically while playing. This helps us better improve your experience.

Improve The AI: If toggled on, the AI will occasionally give you multiple responses to choose from. This helps us evaluate what AI versions are performing better.

Model data, including what was fed into the AI and what the AI output was in turn, will be logged anonymously so we can detect and fix issues to further improve how the AI works.

Banned Words List: The Banned Words list is a way to filter out words that may be sensitive to you. It can be especially helpful in Multiplayer to prevent other players from saying or using specific words. When players try to send actions with Banned Words, the actions will be blocked, deleted, and not used. The Global Banned Words list applies to all Adventures.

The Banned Words List also sends an instruction to the AI to lower the likelihood of using the words when generating responses; however, because of how the AI considers words, the output may still sometimes contain a Banned Word. A potential workaround is to include multiple versions of the same word in the list. For example, if “gun” is added, consider adding “guns” as well.

You can customize multiple aspects of how your AI Dungeon experience looks:

Under this tab, you can:

AI Model: Choose which AI model you’d like to play with. Read more about our different models here.

Story Style: Premium feature. Helps the AI during gameplay with specific genre fine-tune data.

Prompt Settings: Powerful tools for keeping your adventure coherent. Read more about Prompt Settings here.

Presets: Alpha-only feature; adjusts the Advanced Settings to certain groupings.

Advanced Settings: Premium feature. Read more about these options here.

Image Source: Choose between our Pixel Art Image Cache, Fantasy Image Cache, or Stable Diffusion Image Generator.

Image Style: Details for the AI to include across all images you generate in AI Dungeon. Anything in this text box will apply to every picture you add in your story. We have automatically included a few default terms that are appended to your See prompts in an effort to create the best possible output every time.

Image Generation Cost: Starts at one credit and may go up depending on how you adjust the settings to account for increased computation.

Steps: Determines how many passes the model makes during image generation. More steps generally result in higher quality images but also increase generation time and may increase generation cost. Stable Diffusion works well with the default 50 steps. Fewer steps will generate faster results.

CFG Scale: Adjusts how similar your image will be to your prompt. Higher values keep your output closer to your input.

Sampler: The diffusion sampling method, or how it "interprets" the prompt. Each sampler has a different visual impact on the outputs. The default is “k_lms”. Test the same prompt and settings with different samplers to see how they affect your image.

You can customize multiple aspects of how your AI Dungeon experience looks:

Adventure Settings: Options for your current Adventure.

Global Settings: Affects your AI Dungeon experience across every Adventure. These can also be found under your Main Global Settings. Read more about Global Settings here.

Invite Friends or View Players: Play AI Dungeon in Multiplayer mode by inviting friends to your Adventures, then view a list of players once they have joined.