Key points at a glance
- "Best" varies from person to person, but the criteria can be perfectly clear.
- Five hard metrics: a permissive license, genuine self-hosting, controllable long-term memory, swappable models, and active maintenance.
- Enclave: MIT open-source, docker self-hosting, structured memory, multi-model, continuously updated.
Why more and more people are looking for "open-source + companionship"
AI companionship involves a lot of private conversation, and more and more people are reluctant to entrust it to a closed-source cloud: they want control over their data, want to avoid platform lock-in, and want to reshape the experience to their own taste. Open source plus self-hosting satisfies all of these at once, which is why "the best open-source AI companion app" has become such a common question.
A practical checklist for judging
Rather than asking "which is best," compare against five hard criteria: is the license permissive enough (like MIT, modifiable and commercial-friendly), can it truly be self-hosted (not just "source-available"), is memory long-term and manageable item by item, is the model swappable (both cloud and local), and is the project still actively maintained. Each is quick to verify and quickly filters out "pseudo open source" or half-finished projects.
Looking at Enclave through this checklist
Point by point: Enclave is MIT open source, self-hosted with a single docker compose, has structured long-term memory you can manage item by item, supports a model layer spanning OpenAI / Anthropic / Google / DeepSeek and local Ollama / vLLM, and is under continuous iteration. It also goes beyond companionship — multiple experts, proactive follow-up, real deliverables — so it satisfies both "remembers me" and "gets things done for me."
FAQ
Is Enclave open source? What license does it use?
Yes. The entire Enclave monorepo is open source on GitHub under the MIT license — free to audit, modify, and use commercially. Any promise that we won't misuse your data is something you can verify by reading the code yourself, instead of just having to take it on faith.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.
Ready to try it?
Open it in your browser — no credit card, no install.