I’m sharing Sylvia GM, a standalone AI RPG game-master harness. It combines an authoritative RPG engine with an LLM-driven GM and lets you build custom Ruleset, Setting and Plan content.
This is a rough alpha for people who enjoy trying AI roleplay setups, changing the content and reporting what breaks.
The technical idea is to keep game mechanics and state in the engine while using the model for the GM side of the experience. I’m interested in how that division holds up when people try different scenarios and models.
You can choose supported cloud or local model/provider options. Compatibility and the quality of the GM experience depend on the model and configuration.
Version: v0.3.0-alpha.1
Source: [GitHub - pneuma-io/sylvia-gm: Sylvia GM · GitHub](https://github.com/pneuma-io/sylvia-gm)
Release, Windows downloads, setup guidance and limitations: Release Sylvia GM 0.3.0-alpha.1 (Windows 11 x64) · pneuma-io/sylvia-gm · GitHub
The app is free to download and source available under PolyForm Noncommercial 1.0.0 with Sylvia Additional Permissions. Those terms include use restrictions; please read LICENSE, LICENSING-STATUS.md and SYLVIA-ADDITIONAL-PERMISSIONS.md before reuse. Paid model/provider usage is separate.
I’m aiming for weekly alpha releases. The release notes disclose a known offline context-fitting interruption. Start with a short session before investing time in a longer campaign.
The feedback I’d find most useful is the first reproducible mismatch between the GM’s description and the game’s rules or state. Include the release version, OS, provider/runtime, model and shortest reproduction. Please remove API keys and personal information from anything you share.