Key points at a glance
- The point of a self-hosted alternative: data belongs to you, is auditable, and is free from platform lock-in.
- A qualified alternative should be self-hostable, have long-term memory, and offer a swappable model.
- Enclave is MIT open source, docker self-hosted, with structured memory — and it goes beyond Q&A.
Why look for a self-hosted ChatGPT alternative
ChatGPT is powerful, but closed source and cloud-only: conversations sit on the platform, you can't self-host, and you can't personally verify how your data is used. People who want data sovereignty, want to avoid platform lock-in, or want to use it in local and offline environments go looking for an open-source, self-hostable alternative.
What a qualified alternative needs
Three things matter most: it can truly be self-hosted (data on your machine, not just "source-available"), it has long-term memory (remembers you across sessions, rather than starting from zero each time), and the model is swappable (cloud and local alike). Using these three as the bar filters out most solutions that "look the part" but don't work well in practice.
Where Enclave stands
Enclave is MIT open source, self-hosted with a single docker compose, with data stored only locally; it has structured long-term memory, and its model layer supports mixing cloud and local. More importantly, it goes beyond "Q&A" — multiple experts, proactive follow-up, real deliverables — making it a more complete alternative than a single assistant.
FAQ
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.Is my conversation data safe? Will it be used for training?
In self-hosted mode, your conversations live only on your own drive — they never leave your machine and are never used by any third party for training. The model layer can connect to cloud APIs like OpenAI and Anthropic, or be swapped for local Ollama / vLLM to run fully offline; you decide which kind of conversation goes to which model.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.
Ready to try it?
Open it in your browser — no credit card, no install.