Skip to main content
Enclave
Group-chat roleplay

SillyTavern alternative · Group-chat roleplay

Both Enclave and SillyTavern respect open source and self-hosting; the difference is "out-of-the-box" versus "assemble it yourself." SillyTavern is a roleplay chat frontend for tinkerers: you bring your own model backend and manage character cards and prompts, giving you extreme control. Enclave, on the other hand, is a complete world with built-in long-term memory, proactive follow-ups, a network of expert relationships, and real deliverables—ready to use the moment it's installed. If you want full control over prompts and the backend, SillyTavern is powerful; if you want a hassle-free assistant world with memory and deliverables, Enclave is the better fit.

The ceiling of single-character chat

Talk to a single AI long enough and the perspective gets one-note, the rhythm flat, and complex plots or multi-angle debates fall apart. Traditional chatbots also struggle to natively support multiple characters being present at once — they're missing a shared world.

Hand the room over and let the characters tell their own story

In Enclave group chats, every character has their own profile and relationships — both with you and with each other — and they take what others just said into account when they speak. You can play director (set the scene, push the plot) or participant (talk to characters directly), and switch any time.

Side-by-side comparison

DimensionEnclaveSillyTavern
Product positioningAn out-of-the-box private assistant worldA highly customizable roleplay chat frontend; bring your own backend
Open source & self-hostingMIT open source, self-hostableOpen source, self-hostable (self-hosted frontend)
Long-term memoryBuilt-in structured long-term memory, ready out of the boxConfigure it yourself via extensions / vector plugins
ProactivityExperts proactively remind and message you — measured, never spammyPrimarily passive conversation
Multiple characters & relationshipsMultiple experts and a relationship network, with group chats orchestrated by the systemSupports multiple characters, but requires manual orchestration
Real deliverablesCan directly produce PPT / Word / ExcelFocused on roleplay and conversation

How it works in practice

  • Multiple characters present at once

    Pull multiple AI characters into a single group — each keeps their own personality, memory, and relationships with the others.

  • Relationship web between characters

    Friends, rivals, mentors, lovers — the relationships between AIs shape how they interact inside the group.

  • Director mode

    You can quietly push the plot, set the scene, or have a character do something — no need to play the lead in every scene yourself.

  • Shared world state

    Weather, time, and location are shared across the whole group, and the story moves forward along real virtual time.

FAQ

  • Can I use it for TRPGs / murder-mystery games?
    Yes. Every NPC is an independent character with relationships to the others. You GM the scene, and the AIs react based on their own profiles. We're still building deeper rules-system support.
  • Won't multiple characters in one group get chaotic?
    The pace of character messages in a group is scheduled by the system, so they won't flood the chat all at once; you can also remove a character from the group, or take them to a one-on-one chat.
  • Do relationships between AI characters actually change?
    Yes. Events between two characters — group chats they shared, plots where you brought them together or pulled them apart — update their relationship records and shape how they interact next time.
  • Can I self-host Enclave? Is it hard?
    Yes, and it's about as hard as running an ordinary web service. If you've used docker compose, three steps get you your own instance: clone the repo, copy .env, and docker compose up. The API, frontend, database, and vector index all run on your own machine.
  • Which large language models does Enclave support?
    Enclave's model layer is fully swappable: OpenAI, Anthropic, Google, DeepSeek, and local Ollama / vLLM can all be configured. Different characters can even use different models, freely allocated by scenario and cost.
  • What do I need to self-host Enclave? Do I need a GPU?
    Not necessarily a GPU. If you use cloud model APIs (OpenAI, Anthropic, etc.), an ordinary server or home machine that can run Docker is enough — inference happens in the cloud, and your machine only runs the app and database. You only need the corresponding VRAM and compute if you want offline inference with local models (Ollama / vLLM).

Ready to try it?

Open it in your browser — no credit card, no install.

Related

ShareXTelegramLINEWeibo
SillyTavern alternative · Group-chat roleplay · Enclave