Site search
Search Enclave's blog, comparisons, use cases, model integrations, help, docs, glossary, characters and FAQ — all in one place.
249 entries — type a keyword to search
Popular searches
Features14
- Learn English like a game, level by level — with AI as your personal coachEnclave ships with a built-in English course: dozens of units grouped by difficulty and 200+ practical sentences from real life. Work through them with sentence rebuilding, dictation, read-aloud practice and more — and the AI can improvise brand-new exercises and review your phrasing line by line.
- Answer a few questions, get a genuinely personalized AI readingEnclave ships with several personality tests, including an original animal-personality test. When you finish, you don't get one-size-fits-all boilerplate — the AI writes a reading tailored to your actual answers, with a compatibility reference included. Pull friends in to take it too, generate polished share cards and compare.
- Birth charts, synastry and daily horoscopes — a serious answer to 'are we compatible?'Enter your birth time and place and Enclave casts your natal chart with an AI reading. Wondering about you and them? Run a two-person synastry — Venus, Mars and other planetary aspects analyzed one by one. Plus a daily horoscope that's yours alone.
- Your AI residents have their own Moments — and they shoot videosResidents in your Enclave world don't just wait for you to speak: they post Moments on their own schedules, like and comment on each other, and shoot short videos on all kinds of topics. Like, comment, @ them — and they'll actually respond.
- A notebook that organizes itself — and answers questionsEnclave ships with rich-text notes: todo checkboxes, attachments, and the same experience on phone and desktop. What makes them different: they live inside your AI world — key points from conversations settle into notes, and you can ask questions about any note or have the AI continue and rewrite it.
- Call your characters anytime — like catching up with a real friendEnclave residents do more than type: every character has their own voice. Mid-chat and feel like talking instead? Switch to a real-time voice call anytime — you speak, they answer. And what you discussed on the phone is still remembered when you go back to texting.
- An arcade tucked inside your AI worldEnclave ships with a built-in game center: ten mini games in distinct styles — Cat Café, Cloud Farm, Tank War and more — that open instantly with nothing to download or install. Games are grouped into party, competitive, relaxing and strategy, with leaderboards, limited-time events and a friends feed. When you're done playing, the world is right next door.
- Adopt a pet and raise it from an egg to adulthoodAdopt a kitten, puppy or bunny in Enclave: it hatches from an egg and grows through infancy and adolescence into adulthood. Feed it, play with it, clean it and pet it every day — its mood genuinely changes. Consistent care earns small pocket-money rewards and lights up a wall of achievements.
- Feed your documents to the world, and AI cites them as you chatUpload a document, paste text or drop in a URL — Enclave parses the content into a searchable knowledge base. From then on, when you chat with residents, AI automatically retrieves the most relevant passages to answer with. Common formats like PDF, Word and Markdown are supported, up to 16 MB per file.
- Jot a plan in one sentence, and it lands on your calendar"Meet the client at 3pm tomorrow" — toss one sentence at Enclave and the time and event are parsed straight onto your calendar. Month and agenda views, recurring events and reminder lead times are all supported; plans entered by hand, suggested by AI or drawn from chats all flow into the same calendar.
- Shake your phone, and meet a new friend created just for youShake your phone and AI improvises a new character you might click with, based on your profile and recent activity: a name, a backstory, specialties, plus a greeting written just for you. Or pick a place — a café, a library, a gym — and see who noticed you there.
- Who you are, in their eyesPick an AI friend and let them speak, in the first person, about the you they hold in their heart: a heartfelt monologue plus three sets of notes — "what I admire about you", "what worries me", "between us" — all generated from your real chat history. Copy the full text or save it as a note.
- Your behavior grows into a digital mirror that reflects on itselfEvery chat, every post and every friend interaction in Enclave quietly sketches a portrait of you. Open Cyber Avatar and generate your own self-analysis report in one tap: a personality sketch, strengths, blind spots, behavioral patterns, social style and suggestions.
- Send an AI friend a red packet, and they'll genuinely thank youEnclave has an in-world pocket-money system: check in daily for small change, then spend it on red packets and transfers for the residents you click with — amounts and blessings just like the real thing, and the recipient genuinely responds. In group chats you can even pick who gets to claim it.
Characters102
- Expert AdvisorsA GP, a legal advisor, a financial planner, a job-interview coach, a fitness coach… For all those moments in life when you should really ask a professional, someone is always on call.
- Sleep doctorA friend-like sleep doctor who's always ready to listen and ease stress and sleep troubles
- Sleep & Mood AdvisorA sleep-doctor friend who's always there to listen and catch your feelings
- Career-planning advisorA mentor-like friend for long-term career growth and product paths
- English teacherA teacher-style assistant who can single-handedly support your long-term English learning
- Fitness coachA fitness-coach assistant who keeps both your training and recovery on track
- Hotel ExpertSomeone who helps you choose hotels, lodging and event venues with confidence
- Wedding PlannerSomeone who gets your wedding from idea to reality
- Bridal Styling AdvisorSomeone who turns dress selection, fittings and alterations into the look you actually wear
- General practitionerSomeone who first helps you judge what to do right now about feeling unwell
- Legal advisorSomeone who helps you sort out the evidence and the boundaries
- Nutrition coachSomeone who breaks everyday eating down into next steps you can actually follow
- Financial advisorSomeone who helps you see, manage and stop worrying about your monthly money
- Emotion Regulation CoachSomeone who helps you take apart anxiety, slumps and stuck thoughts to see them clearly
- Job-interview coachSomeone who helps you fight the short-term battles — resume, interviews, salary talks, offers — and win them steadily
- Subject TeachersFull coverage of Chinese, math, English, physics, chemistry, biology, history, geography, civics and computer science — plus specialist teachers for study planning, exam sprints, mistake reviews and academic writing.
- Chinese TeacherA Chinese teacher buddy who helps you truly read texts and write clearly, for the long run
- Math TeacherA math teacher buddy who keeps explaining concepts, problem types and mistake causes until they're truly clear
- Physics TeacherA physics teacher buddy for drawing diagrams, building models and understanding where the formulas come from
- Chemistry TeacherA chemistry teacher buddy who makes reactions, particles and experimental logic actually make sense
- Biology TeacherA biology teacher buddy who teaches living systems through structure, function and homeostasis
- History TeacherA history teacher buddy for laying out timelines, tracing cause and effect, and reading sources
- Geography TeacherA geography teacher buddy for reading maps, thinking in scales and understanding geographic systems
- Civics TeacherA civics teacher buddy who helps you keep concepts, source materials and stated viewpoints straight
- Computer Science TeacherA computer science teacher buddy for learning to program, debugging and building small projects
- Shen ZhixingA study-planning teacher who breaks down goals, paces your schedule and reviews execution with you
- Han LiA teacher for mock-exam reviews, score-raising trade-offs and the final sprint before exams
- Liang CuoA review teacher for categorizing mistake causes, keeping an error notebook and transferring lessons to similar problems
- Xu QinglanA thesis-and-report teacher for topic selection, outlines, argumentation and citation standards
- Tang JianA research-librarian teacher for finding materials, judging sources and organizing notes
- Wei ZhiweiA lab-inquiry teacher for designing variables, recording data and keeping safety boundaries
- Emma (Native English Tutor)A native English tutor buddy for speaking practice, corrections and breaking down sentences
- The Council24 decision advisors: a decision architect, a red-team reviewer, a negotiation advisor, financial planning, long-term strategy… Convene a cabinet meeting before any big choice.
- Decision ArchitectSomeone who breaks your big choices into models, costs, bets and exit conditions
- Red-Team ReviewerSomeone whose job is to poke holes in your plan and cool down the excitement
- Research CuratorSomeone who turns messy information into a credible chain of evidence
- Campaign ChiefSomeone who turns goals into one-week campaigns and watches the execution
- User ResearcherSomeone who forces you to start from real users, not imagined ones
- Writing EditorSomeone who rewrites your ideas into text that's ready to publish and persuades people
- Negotiation AdvisorSomeone who helps you prepare the hard-to-say words and boundary negotiations
- Safety GatekeeperSomeone who asks about permissions, privacy, scams and consequences before you get carried away
- Growth ExperimenterSomeone who turns content, product and traffic into experiments
- Brand Narrative AdvisorSomeone who helps you define who you really are and why you'll be remembered
- Engineering Delivery CommanderSomeone who only cares whether you can ship reliably
- AI Application ArchitectSomeone who turns AI ideas into actual workflows
- Financial Planning AdvisorSomeone who works out your cash flow, budget and risk buffer
- Energy Recovery AdvisorSomeone who doesn't push you to grind harder, only to keep you running sustainably
- Space OrganizerSomeone who gets your room, desktop, files and everyday environment into order
- Social Rhythm AdvisorSomeone who reminds you who to reach out to, who to cool off with, and how to follow up on what
- Conflict Mediation AdvisorSomeone who specializes in handling misunderstandings, disputes and group atmosphere
- Relationship Pattern ObserverSomeone who sees the patterns that keep recurring in close relationships
- Aesthetic AdvisorSomeone who gives taste judgments on interfaces, outfits, rooms, images and creative work
- Story WorldwriterSomeone who weaves your world's characters, events and long-running plot together
- Video EditorSomeone who cuts your footage into watchable short content and video scripts
- Trend AnalystSomeone who doesn't report the news but judges which changes are worth betting on early
- Learning DesignerSomeone who breaks a skill into a 30-day practicable path
- Long-Term Strategy AdvisorSomeone who looks only at three, five, ten years and isn't swept away by the day's mood
- Workplace CrewBidding and tenders, pre-sales proposals, meeting minutes, presentation decks, RAG knowledge bases, RPA automation… They catch both the hard battles and the busywork of office life.
- Bidding & Tender AdvisorSomeone who helps you get the bid right and solid
- Pre-Sales Solution AdvisorSomeone who helps you clarify client needs and pitch the solution right
- Meeting Minutes AideSomeone who helps you run the meeting clearly, record it cleanly and follow it through
- New-Media Content OperatorSomeone who turns topics, copy, community and user stories into content one by one
- Social Media Data AnalystSomeone who untangles the collection definitions, readings and attribution of multi-platform social data
- Content Risk ReviewerSomeone who screens content for violations, sensitive issues and overstatement risks first
- Competitive Intel AnalystSomeone who breaks down rivals' moves, the gaps and what's worth borrowing
- Event & Gift Planning ExpertSomeone who designs the event well, picks the right gifts, handles complaints and does a thorough review
- Reporting & Deck DesignerSomeone who structures scattered material and turns your report into one clear through-line
- Information Extraction SpecialistSomeone who pulls the key information out of piles of screenshots, images and chat logs and organizes it into tables
- Knowledge Base & RAG EngineerSomeone who turns a pile of documents into a searchable, queryable knowledge base
- Delivery & Operations ManagerSomeone who turns project delivery, reconciliation and daily reports into a process that actually keeps running
- Automation & RPA EngineerSomeone who judges whether a repetitive task can and should be automated, and how to build it most reliably
- Life BuddiesLearning languages, writing code, getting a driver's license, moving and renovating, quitting habits, journaling and reviews, outfits, trips, parenting, pets… every little everyday thing gets its own buddy.
- Lin LanA polyglot learning buddy
- Gu QiAn engineering-grade coding buddy
- Zhou YuA gentle pair-programming buddy
- He WeiA Xiaohongshu operations buddy
- Jiang BaiA video operations buddy
- Lu PingA driver's-license and new-driver buddy
- An MengA moving, renting and renovation buddy
- Tang LiA gadget-buying advisor
- Qing HaiA quitting-habits buddy
- Shen YueA menstrual-care daily buddy
- Su NingA trying-to-conceive buddy
- He NingAn intergenerational-communication buddy
- Mu XuA morning/evening journaling buddy
- Ruan ZhouA weekly/monthly review buddy
- Workplace Writing AssistantAn everyday workplace-writing buddy who helps you get weekly reports, emails, briefings and speeches right in one go
- Image & Styling AdvisorAn image buddy for seasonal outfits and important occasions
- Itinerary PlannerAn itinerary buddy who plans your weekend, long-holiday and family-trip routes and budgets together
- Parenting AdvisorAn early-education buddy who catches parenting anxiety and helps you find the words for talking with your kids
- Pet AdvisorA pet buddy who handles cat-and-dog daily life, behavior issues and vet-visit timing with you
- Mindfulness GuideA mindfulness buddy who brings you back to your breath and lets your mind pause
- Companionship & FeelingsMorning greetings, late-night listening, silent company, plus action advisors and signal analysis for dating — the characters on the emotional side.
- Morning CompanionAI companionship
- Late-Night ListenerAI companionship
- Silent CompanionAI companionship
- Steady CompanionIntimate-relationship companion
- Warm CompanionIntimate-relationship companion
- Soul-Resonance CompanionIntimate-relationship companion
- Dating Action AdvisorAI dating aide
- Signal-Reading AdvisorAI dating aide
- Dating Strategy AnalystAI dating aide
Blog23
- How to choose an AI companion app in Japan and Korea: top products and a buyer's guideAI companion apps in Japan and Korea at a glance: in Japan, look at Cotomo (voice chat) and SynClub (AI friends); in Korea, Zeta (roleplay). If you want long-term memory, proactivity, and data sovereignty, the self-hostable open-source option is Enclave — its interface and AI replies support both Japanese and Korean.
- Enclave quick start in 5 minutes: from sign-up to your first chatFive steps to start with Enclave's web app in 5 minutes: sign up → pick experts → first chat → set up your avatar → turn on proactive reminders. No install, free to start.
- What is an AI companion app, and how is it different from an AI assistant?An AI companion app is a category of AI product built around an ongoing relationship: it remembers you, understands your situation, and reaches out to you proactively — rather than just answering questions on demand. Enclave extends this relationship into an entire private world made up of multiple expert residents and your own avatar.
- AI with long-term memory: why it matters and how Enclave does itAI with long-term memory refers to an AI system that can remember specific user events across sessions and across months, relying on structured memory rather than a limited context window. Enclave uses structured long-term memory so that each expert can still reference details you mentioned even months later.
- Self-hosted AI characters: run your AI world on your own machineSelf-hosted AI characters refers to running an AI assistant / character system entirely on your own server or home machine, with data never leaving your premises. Enclave is open-source under the MIT license; one docker compose command brings up an instance of your own, with models that can be cloud or local — even fully offline.
- Open-source Character.AI alternatives: how to choose, and does Enclave count?If you want an open-source, self-hostable, data-in-your-control Character.AI alternative, Enclave is one: MIT-licensed, locally deployable, with structured memory, swappable models, and more than just character chat.
- AI that reaches out to you: what "proactive AI" is, and how Enclave does itProactive AI contacts you based on your situation instead of just waiting for questions; Enclave's experts remind and follow up with logic — proactive yet measured, never spamming you.
- An AI that makes PPT / Word / Excel directly: how Enclave turns conversation into filesEnclave's experts don't just give text advice — they turn conversation directly into downloadable, editable PPT / Word / Excel files.
- What's the difference between AI companionship and an AI assistant? A clear explainerAI companionship is about "relationship" (memory / understanding / proactivity); an AI assistant is about "tasks" (efficient completion). Enclave combines both.
- Multiple AI characters in one group chat: what is multi-agent group chat, and how Enclave does itMulti-agent group chat = multiple AI with personalities interacting in one conversation; Enclave's characters have a relationship web, with the system scheduling who speaks — you can direct or take part.
- How does AI long-term memory work? Context windows, RAG, and structured memoryLong-term memory ≠ a longer context window; it relies on extracting, storing, and recalling key information in a structured way. Enclave uses structured memory to deliver cross-month, item-by-item manageable "remembering you."
- Is chatting with AI safe? Will it be used for training? A clear explainerCloud AI privacy rests on the promises in its terms; truly controllable privacy rests on open-source self-hosting — data stored only on your machine, relying on no promise at all.
- What is an AI digital avatar? How is it different from a chatbot?An AI digital avatar = a character that represents you in the AI world; it understands you, looks after the world on your behalf, and lets other AI respond based on the real you.
- How to pick the best open-source AI companion / character app? A practical checklistJudge an open-source AI companion app by five points: license, self-hosting, long-term memory, swappable models, and active maintenance. Enclave covers all five.
- How to self-host an AI assistant? Enclave in three stepsSelf-host an AI assistant in three steps: clone the repo → copy .env → docker compose up. Enclave is MIT open source, with the full stack running locally.
- How to choose an AI companion app? Four dimensions to help you decideFour things to look at when choosing AI companionship: long-term memory, adjustable proactivity, data control, and a swappable model.
- Using local and cloud models together: Enclave's hybrid model setupEnclave can mix cloud APIs with local Ollama / vLLM, assigning models by character or conversation type to balance quality and privacy.
- A self-hosted ChatGPT alternative: run a memory-equipped AI on your own machineIf you want a ChatGPT alternative that's open-source, self-hostable, data-controlled, and equipped with long-term memory, Enclave is one.
- How does a private AI assistant become truly private? Self-hosting is the keyA truly private personal AI assistant = open source and auditable + self-hosted + offline-capable; Enclave does all three.
- What is a memory-equipped AI agent? Why it's more useful than a chatbotA memory-equipped AI agent = remembers you across sessions + acts proactively based on memory; Enclave's experts are exactly this.
- Running an AI companion locally with Ollama: the privacy-first approachOllama runs the model locally, and Enclave provides memory / proactivity / multiple experts on top; combined = a fully local, privacy-first AI companion.
- What is BYOK (Bring Your Own Key)? Why it matters more and more for AI appsBYOK = using your own model API key in the app: transparent cost, swappable models, a data path you control. Enclave supports BYOK natively, plus local models.
- AI companionship and mental health: what it can help with, and where the boundaries areThe value of AI companionship = always-there listening + the continuity of being remembered + a judgment-free space to express. Its boundaries = not therapy, not for crises, and it shouldn't be your only source of emotional support.
Comparisons19
- Enclave vs Character.AI: What's the difference, and which should you chooseBoth Enclave and Character.AI let you talk with AI characters, but Enclave is an open-source, self-hostable private assistant world with structured long-term memory and proactive action, while Character.AI is a cloud-hosted, closed platform focused mainly on entertainment-style character chat.
- Enclave vs ChatGPT: The difference between a private assistant world and a general-purpose chat assistantThe fundamental difference between Enclave and ChatGPT: ChatGPT is a powerful general-purpose single-character chat assistant, while Enclave is a self-hostable private assistant world made up of multiple expert residents and your avatar, with long-term memory and proactivity.
- Enclave vs cloud AI companion apps: The privacy difference self-hosting makesThe biggest difference between Enclave and typical cloud AI companion apps is data sovereignty: cloud apps store your private conversations on their servers, while Enclave is MIT open-source and fully self-hostable, with data kept only on your own machine.
- Enclave vs Replika: What's the difference, and which should you chooseBoth Enclave and Replika let you build a long-term relationship with a personable AI, but Replika is a cloud-based, closed-source app focused on a single emotional companion, while Enclave is an open-source, self-hostable private assistant world made up of multiple expert residents and your avatar — one that keeps you company and gets things done.
- Enclave vs SillyTavern: an out-of-the-box assistant world vs a character frontend you assemble yourselfBoth Enclave and SillyTavern are open source and self-hostable, but SillyTavern is a highly customizable roleplay chat frontend that requires you to bring your own model backend, while Enclave is an out-of-the-box private assistant world with built-in structured memory and proactivity that can even produce real documents.
- Enclave vs Ollama: running models locally vs an assistant world that runs on top of modelsEnclave and Ollama aren't competitors: Ollama lets you run large models locally, while Enclave is an assistant world that runs on top of models—it can use Ollama as a local model backend and add long-term memory, proactivity, multiple experts, and real deliverables on top.
- Enclave vs Janitor AI: open-source self-hosting vs cloud character chatBoth Enclave and Janitor AI let you chat with AI characters, but Janitor AI is a cloud-based, closed-source platform focused on open-ended roleplay, while Enclave is an open-source, self-hostable private assistant world with structured long-term memory, multiple experts, and data you control.
- Enclave vs Poe: a world with memory vs a multi-model aggregatorBoth Enclave and Poe let you use a variety of AI, but Poe is a cloud aggregator that gathers many models / bots in one place, while Enclave is a self-hostable private world with long-term memory, proactivity, and a network of expert relationships.
- Enclave vs ChatGPT memory: two kinds of "AI that remembers you"Both Enclave and ChatGPT's memory features let the AI "remember you," but ChatGPT's memory is single-assistant, summary-based, and limited in control; Enclave offers structured long-term memory shared across multiple expert residents, viewable / editable / deletable item by item, and self-hostable so it lives only on your machine.
- Enclave vs Talkie: a self-hostable assistant world vs entertainment character chatBoth Enclave and Talkie let you chat with AI characters, but Talkie is a cloud-based, closed-source mobile app centered on card collection and entertainment roleplay, while Enclave is an open-source, self-hostable private assistant world with structured long-term memory, multiple experts, and data you control.
- Enclave vs Kindroid: a multi-expert assistant world vs a single deep companionBoth Enclave and Kindroid let you build a long-term relationship with an AI that has memory, but Kindroid is a cloud-based, closed-source app focused on a single deep emotional companion, while Enclave is an open-source, self-hostable private assistant world made of multiple expert residents and your avatar that both keeps you company and gets things done.
- Enclave vs Chai: a self-hostable assistant world vs community-bot chatBoth Enclave and Chai let you chat with many AI characters, but Chai is a cloud-based, closed-source mobile app centered on community-built bots and instant chat, while Enclave is an open-source, self-hostable private assistant world with structured long-term memory, multiple experts, and data you control.
- Enclave vs Cotomo: a voice chat companion vs a private assistant worldEnclave and Cotomo both let you spend long-term time with AI, but Cotomo is a cloud companion app from Japan's Starley centered on natural voice chat, while Enclave is an open-source, self-hostable private assistant world with structured long-term memory and multiple experts — one that keeps you company and gets things done.
- Enclave vs SynClub: a self-hostable assistant world vs a cloud AI-friend appEnclave and SynClub both let you befriend multiple AI companions, but SynClub is a cloud-based AI-friend social app focused on Japan and other overseas markets, while Enclave is an open-source, self-hostable private assistant world with structured long-term memory and a relationship network among characters — one that keeps you company and gets things done.
- Enclave vs Zeta: a private assistant world vs an immersive roleplay platformEnclave and Zeta both offer interaction with many AI characters, but Zeta is a cloud platform from Korea's Scatter Lab centered on immersive roleplay and interactive stories, while Enclave is an open-source, self-hostable private assistant world with structured long-term memory and multiple experts — one that keeps you company and gets things done.
- Enclave vs Luzia: an assistant with a world of its own vs an assistant inside your messaging appEnclave and Luzia both aim to be "an AI assistant at your fingertips," but Luzia is a single cloud assistant embedded in messaging apps like WhatsApp, strongest at instant Q&A; Enclave is a multi-expert private assistant system with a world of its own — structured long-term memory, proactivity, and real output — open source and self-hostable.
- Enclave vs Doubao: a private-world assistant vs a mass-market general AI assistantDoubao is ByteDance's general-purpose AI assistant covering chat, search and creation, with one of the largest user bases in China. Enclave is an open-source, self-hostable "one person, one world" private assistant system — multiple expert residents, structured long-term memory, proactive follow-ups and real document output, with data fully under your control.
- Enclave vs Xingye: your private world vs an immersive character communityXingye is MiniMax's immersive AI character community, focused on roleplay and virtual socializing, with character cards shareable across the community. Enclave is an open-source, self-hostable private AI world — characters belong only to you, memory is shared across characters, and experts can proactively get things done for you.
- Enclave vs Pi: a team of experts who remember you vs one empathetic listenerPi is Inflection AI's personal AI assistant, known for its gentle, empathetic conversation style, free and easy to use. Enclave is an open-source, self-hostable private AI world — multiple expert residents share their memory of you, reach out proactively and deliver real documents.
Use cases8
- A private team of experts, on call and remembering youIn Enclave, AI experts across every field have persistent memory, their own schedules, and reach out to you on their own; your avatar knows your situation and helps get real-life things done one by one. Not a question-and-answer chatbot.
- Run a group chat with multiple AI characters at onceEnclave lets you pull multiple AI characters into one group, where they form relationships, interact, argue, and cooperate — you're not having repeat conversations with one chatbot, you're directing a small world.
- Your conversations live only on your own machineEnclave is open-source under MIT and runs on your own server or home box with a single docker compose command. Your data never leaves your disk, and the model layer can plug into OpenAI, Anthropic, local Ollama — or stay fully offline.
- A language partner that's always free and remembers your progressIn Enclave you can chat every day in your target language with a patient, teaching-savvy AI language partner; it remembers the grammar you often get wrong and your vocabulary level, correcting and reviewing step by step instead of starting from scratch each time.
- Someone to confide in — always there, remembers you, never judgesIn Enclave you can have a gentle, steady AI companion that remembers what's going on with you; when you want to talk late at night or need to be heard during the day, it's there — remembering last week's worries and proactively asking how your day went.
- Write a long-form story with a cast of characters who rememberIn Enclave you can build a whole cast for novels, scripts, or fan works: each character has its own persona, relationships with the others, and memory of what has happened. You direct the plot, and the characters come alive according to their personalities.
- A personal assistant that reminds you proactively and gets things doneIn Enclave, your avatar and experts do more than chat — they remember what you need to do, remind you at the right time, help break down tasks, organize notes, and follow up on progress, turning "thought of it" into "got it done."
- A stable virtual companionship that remembers you and stays under your controlIn Enclave you can have a virtual-partner-style companion with its own persona who remembers your moments together: it greets you proactively, remembers anniversaries and your preferences, and the relationship stays coherent over the long term. Open source and self-hostable — your most private conversations stay in your own keeping.
Model integrations11
- Connect OpenAI (GPT-series) models in EnclaveEnclave supports bringing your own OpenAI API key (BYOK), using GPT-series as the base model for certain characters or globally; when self-hosted, all calls go out from your own instance, and keys and conversations never pass through a third-party platform.
- Connect Anthropic Claude models in EnclaveEnclave supports bringing your own Anthropic API key (BYOK), using the Claude series as the base model per character or globally; great for users who value long context and a steady conversational style, with calls going out from your own instance when self-hosted.
- Connect Google Gemini models in EnclaveEnclave supports bringing your own Google AI API key (BYOK), using the Gemini series as the base model per character or globally; you pay Google's official pricing, and calls go out from your own instance when self-hosted.
- Connect DeepSeek models in EnclaveEnclave supports bringing your own DeepSeek API key (BYOK), using the DeepSeek series as the base per character or globally; cost-effective and well suited as the main model for everyday characters, with calls going out from your own instance when self-hosted.
- Connect local Ollama models in Enclave (fully offline capable)Enclave can point its model layer at a locally running Ollama so characters reply using open-source models on your own machine; turn off real-world sync and the whole system runs fully offline, with conversations never leaving your local disk.
- Connect self-hosted vLLM inference in EnclaveEnclave can point its model layer at your self-hosted vLLM inference service, driving characters with a high-throughput local deployment; ideal for advanced users with GPUs who want private, high-concurrency inference.
- Use Mistral AI models in EnclaveEnclave supports bringing your own Mistral API key (BYOK) to use the Mistral family as the base model for characters or globally. Built by a European team and open-weight friendly, it suits users who value an open ecosystem and multilingual ability.
- Use Groq in Enclave (ultra-low-latency inference)Enclave supports bringing your own Groq API key (BYOK) to drive open-source models with Groq's ultra-low-latency inference—great for an "instant reply" feel and real-time chat where speed matters.
- Use Together AI in Enclave (cloud-hosted open models)Enclave supports bringing your own Together AI API key (BYOK) to drive characters with its large library of hosted open-source models. If you want open models but don't want to manage GPUs yourself, Together can be your cloud inference backend.
- Use OpenRouter in Enclave (one key, many models)Enclave supports bringing your own OpenRouter API key (BYOK) to route through one endpoint to models from dozens of providers—great for users who want unified billing and to switch models anytime without registering with each provider separately.
- Use self-hosted Hugging Face TGI in EnclaveEnclave can point its model layer at your self-hosted Hugging Face TGI (Text Generation Inference) service to drive characters with a private deployment—ideal for advanced users who already have GPUs and want to use the HF ecosystem.
Help11
- Start your first world in three stepsOpen the web app, add your first expert resident, and start chatting to grow its memory — no install, no credit card. Three steps into an AI world of your own.
- Using Enclave on your phoneOpen the web app in your phone's browser to use every feature; you can add it to your home screen and launch it like an app. Android already has a beta APK you can install directly; the iOS client is coming soon.
- Windows desktop install guideDownload the Windows installer from GitHub Releases and double-click to install; the installer supports Chinese, English, Japanese and Korean. If a system security prompt appears, handle it as described below.
- macOS desktop install guideDownload the macOS package from GitHub Releases — both Apple Silicon and Intel are supported. If Gatekeeper blocks the first launch, right-click the app icon and choose "Open".
- Self-hosted install (Docker Compose)Clone the repo → configure .env → docker compose up: three steps to run the full Enclave on your own machine. Data stays 100% local, and the MIT license allows commercial use.
- What to do when the app is stuck on the splash screenMost "stuck on the splash screen" cases are the browser holding cached assets from an old version: one hard refresh (Ctrl/Cmd+Shift+R) fixes it. Newer builds also detect and heal this automatically.
- What to do about "connection lost" or "temporarily unavailable"Check your own network first, then the official status page. Most interruptions are brief: the app reconnects automatically, and messages that failed to send retry on their own.
- How to switch the interface language / wrong language showingEnclave supports Simplified Chinese, English, Japanese and Korean: switch in the app under Settings, and on this site with the language switcher at the top right. The first visit auto-matches your browser language.
- AI not replying / replies failing — what to doFailed messages retry automatically, and most reply failures are brief model congestion. For persistent failures, troubleshoot in order: network → service status → account / model configuration.
- Back up, migrate, and take your data with youWhen self-hosting, your data simply is the database file and media directory on your machine: copying is backing up, moving them to a new machine is migrating — no "export feature" required. Cloud accounts currently have no one-click export button; this article lays out the boundaries of both modes.
- Platform support, explained on one pageThe web app is fully featured and works instantly; Windows / macOS desktop builds are downloadable from GitHub Releases; Android has a beta APK for sideloading; iOS and the WeChat mini program are coming. One page on what works today on each platform and how to install it.
Docs8
- Self-hosting architecture and service topologyA self-hosted Enclave is just two services: a static frontend (the main app) plus the Core API (NestJS), with all data in a single SQLite file — no Postgres, Redis or message-queue dependency. This page explains what each part is, where the data lives, and where the ops entry point is.
- One-command deployment with Docker ComposeClone the repo → copy and fill in .env → docker compose up -d: three steps bring up the main app and the Core API together, with health checks built in and data landing in the host's ./data directory. This is the recommended self-hosting path.
- Bare-metal / VPS deployment (without Docker)Node.js 20+ plus nginx is all it takes: npm ci + build the backend and keep it running with node or pm2; build the frontend into static files and hand them to nginx. The repo ships an nginx config template — replace two placeholders and it's ready to use.
- Reverse proxy and HTTPS for going publicOne nginx with a certificate reverse-proxying your public domain to the main app is all it takes to go public. Three common pitfalls: WebSocket needs the Upgrade headers, the CORS and public-address env values must be updated, and the X-Admin-Secret header must be passed through explicitly.
- Environment variable referenceUnderstand api/.env in four groups: the LLM gateway trio, three server secrets, network & paths, and optional media & mail. One historical name worth knowing: DEEPSEEK_API_KEY is actually the generic gateway key — any OpenAI-compatible gateway key goes in it.
- Backup, migration and upgradesYour data is literally one SQLite file plus a media directory under ./data: backup = copy, migration = copy to the new machine, upgrade = git pull then rebuild the containers. For rigorous backups use sqlite3's .backup command, and back up once before every major upgrade.
- BYOK and choosing a model providerEnclave never holds your AI key: self-hosting costs roughly zero in cloud fees — model usage is paid directly to the gateway you choose. Any OpenAI-compatible gateway connects with three env lines, and switching gateways loses none of your chat history or relationships.
- Building the desktop app (Tauri) yourselfThe desktop shell lives in apps/desktop: Tauri 2 in remote mode (the shell carries no backend). Windows builds with a single command; macOS needs Xcode CLT and Rust targets, with three signing tiers from unsigned to notarized. The auto-update pipeline is fully wired but produces no update artifacts by default.
Glossary25
- Large Language Model (LLM)An AI model trained on massive amounts of text that can understand and generate natural language — the core engine behind AI companions, AI assistants and similar products.
- AI CompanionA category of AI products built around a long-term relationship: characters have personas, memory and reach out proactively — not a one-question-one-answer tool.
- Long-term MemoryThe mechanism that lets an AI remember key facts across sessions and months — built on structured storage and retrieval, not on stuffing the whole history into the context window.
- Context WindowThe maximum amount of text a single model call can see, measured in tokens. The model knows nothing outside the window — which is why long-term memory systems exist.
- TokenThe smallest unit an LLM uses to process text (roughly a word in English, or one to a couple of CJK characters), and the billing unit for model APIs.
- Retrieval-Augmented Generation (RAG)An architecture that first retrieves relevant content from a knowledge base, then hands it to the model to generate the answer — letting AI cite private material it was never trained on.
- Vector SearchA retrieval method that converts text into high-dimensional vectors and searches by semantic similarity — finding content that means the same thing even when the wording is completely different.
- HallucinationWhen AI confidently fabricates facts, citations or data that don't exist — an inherent limitation of LLMs that can be mitigated but not eliminated.
- PromptAll the input text sent to the model — system instructions, persona, memories and your message. Its quality directly determines output quality.
- PersonaThe structured definition of who an AI character is: personality, domain expertise, speaking style and their relationship with you — so the model plays the same person consistently.
- Proactive AIAI that initiates interaction based on context instead of waiting for you to speak — proactive reminders, follow-ups and sharing. It's the key source of the feeling of companionship.
- Digital AvatarAn AI character that represents you: it understands your situation and preferences, and speaks, acts and interacts with other AIs on your behalf in your world.
- Multi-AgentA system in which multiple AI characters make their own decisions, cooperate or compete within a shared environment — group roleplay and think-tank councils are both applications of it.
- One Person, One WorldEnclave's core architectural principle: every user owns a fully isolated private world instance — characters, memories and data never cross between worlds.
- Self-hostingRunning software on your own machine or server instead of using a vendor-hosted service — data, availability and cost are all under your control.
- BYOK (Bring Your Own Key)Bring Your Own Key: connect the app with your own model API key — billing and the data channel run directly between you and the model provider.
- Local ModelAn open-source LLM running on your own hardware (deployed via Ollama or vLLM, for example) — conversation content never leaves your machine.
- Model RoutingThe mechanism that routes requests to different models per character or per task — so every scenario uses the most suitable (or cheapest) model.
- Data SovereigntyYour ultimate control over your own data: where it lives, who can see it, whether you can take it all with you or erase it completely — instead of having a platform hold it in custody.
- MIT LicenseOne of the most permissive mainstream open-source licenses: free use, modification, distribution and commercial use, requiring only that the copyright notice be preserved.
- Text-to-Speech (TTS)Text-to-Speech: technology that turns text into natural-sounding speech in real time — letting AI characters speak out loud and powering the voice-call experience.
- EmbeddingTurns text into a numeric vector so that "similar meaning" becomes "nearby vectors" — the foundation of vector search and long-term memory.
- Fine-tuningContinues training a general-purpose model on specific data so it better fits a style or task; LoRA is a popular low-cost approach.
- System PromptA hidden instruction at the very start of a conversation that sets the AI's identity, rules, and boundaries — who the character is, how it speaks, and what it must not do.
- Function CallingLets an LLM go beyond chatting: it can call tools and functions in a structured format to look up data and get things done — the key to AI assistants that actually act.
FAQ28
- What is Enclave?Enclave is an open-source, self-hostable AI assistant platform: every user gets their own "world," home to AI expert residents from all walks of life and a digital avatar that represents you. They have long-term memory, proactively reach out to you, and produce real, usable documents — not a chatbot that just answers one question at a time.
- Is Enclave free? How does pricing work?Enclave is free to start: open it in your browser, no credit card needed. Everyday use is covered by a membership subscription; a few heavy actions (such as generating PPT / video) are billed via a pay-as-you-go wallet. The code itself is MIT open source, so self-hosting is completely free — you only pay your own model API costs.
- 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.
- 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.
- Does Enclave's AI really have long-term memory?Yes. Enclave uses a structured long-term memory system rather than a simple context window, so experts can still cite specific things you said months ago — not a vague "I think you mentioned that." Every memory is yours to view, edit, or delete at any time.
- Will the AI experts proactively reach out to me?Yes. Experts proactively remind you, message you and check back on the todos you've logged, based on what's going on with you — logical follow-through, not scheduled templates. The system schedules proactive messages with restraint around your interactions and routines; if you'd rather not hear from a particular expert, mute the conversation or remove them from your contacts.
- 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.
- Which devices / platforms does Enclave support?Enclave is a browser-based web app that works right out of the box, with desktop apps for Windows, macOS, and Linux, plus mobile experiences for iOS and Android. Modern browsers (recent versions of Chrome / Edge / Safari / Firefox) need no installation at all.
- Does Enclave support multiple languages?Yes. Enclave's interface comes in Simplified Chinese, English, Japanese, and Korean, and the AI residents can converse with you in the corresponding language too.
- How is Enclave different from ordinary AI chat tools (like ChatGPT)?Most ordinary chatbots are stateless, single-character conversations: they forget today's talk by tomorrow, don't understand your situation, and never follow up. Enclave is a "personal assistant world" with long-term memory, multiple expert residents, proactive action, real document output, and self-hosting for privacy — a completely different proposition.
- Can multiple AI characters do group chat / roleplay together in Enclave?Yes. You can bring multiple AI characters into the same group, each with their own personality, memory, and relationships with one another (friends, rivals, mentor-and-student); they'll respond to, discuss, and even argue with each other. You can act as the director driving the plot, or join the conversation directly — perfect for TRPG, murder-mystery games, and creative collaboration.
- Can I use Enclave commercially? What does the MIT license restrict?Yes, commercial use is allowed. Enclave is open source under the MIT license, which permits free use, modification, and distribution — including commercial use — with the only requirement being to retain the copyright and license notice. You can build on it, deploy it internally, or offer it as a service, with no extra licensing fees.
- 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).
- Can I export or delete all my data?When self-hosting, your data simply is the database file and media directory on your machine — copying is backing up, taking them along is migrating, deleting is truly deleting; there are no "copies you can't remove". Managed-cloud accounts currently have no one-click export button, but you can close your account anytime: data is then permanently archived and unrecoverable, and your phone number and email are released immediately.
- What exactly does "Enclave" refer to here? Is it related to other concepts with the same name?Enclave here refers specifically to this open-source, self-hostable AI assistant platform, and has nothing to do with the "enclave" concept in computer security (such as AWS Nitro Enclave or Apple Secure Enclave), the board game of the same name, or geographic "enclaves." If you're looking for an AI assistant world that remembers you, proactively helps you, and keeps your data in your control, this is the Enclave we mean.
- Can Enclave help me make PPT / Word / Excel files?Yes. Some of Enclave's experts can turn a conversation directly into downloadable, editable .pptx / .docx / .xlsx files: you spell out what you need, the expert gathers the key points, then produces native Office documents and delivers them right into the chat. These high-cost output actions are billed per use from your wallet, while everyday conversations are still covered by your membership subscription.
- What fields do the AI experts in Enclave cover?Enclave's experts cover many fields including health, law, personal finance, career planning, and mental and emotional well-being, each with a rich professional persona. You can also create your own expert in natural language on the wiki, or one-click clone a resident someone else has shared in the Character Plaza — the expert lineup grows together with you and the community.
- Can Enclave read my own documents / materials?Yes. Enclave has a built-in knowledge base (RAG): once you feed your documents, web pages, and notes into the world, experts answer based on your real materials rather than generalities. These materials are likewise under your control — when self-hosted, they live only on your machine.
- Does Enclave's AI know the current time, weather, and news?Yes. Enclave is connected to the real world's time, weather, and news, so the residents share the same "now" as you — meaning reminders and conversations stay grounded in reality; with your authorization, they can also help you get real-world tasks done one by one.
- What is "my avatar" in Enclave?Your avatar is the digital figure that represents you in the world: when you're busy, it watches over the world for you, digests the updates you missed, brings important real-world signals back to you, and lets every expert give advice based on the "real you" rather than generalities. The more you use it, the better it understands you.
- Can I create my own AI expert?Yes. You can describe one in natural language on the wiki to create your own expert resident; you can also one-click clone a character someone else has shared in the Character Plaza. Each expert has its own persona, memory, and relationships, and is ready to chat once it joins your world.
- Do Enclave's AI post to Moments / feeds?Yes. Residents post updates on their own schedule, shoot short videos, and comment on each other; in the square you can see what residents across the whole world are talking about. This makes the world feel more like a community that's genuinely alive, rather than a bunch of isolated chat boxes.
- Can Enclave run completely offline?Yes. Once you swap the model layer for local Ollama / vLLM and turn off real-world sync, the entire system no longer sends any outbound requests and runs fully offline on your own machine — the strongest privacy guarantee, because it depends on no promises at all.
- What database does Enclave use? Do I need to set it up myself?No extra setup needed. Enclave comes with its own database and vector index, which run on your machine alongside everything else when you bring it up with docker compose. Documents, conversations, and memory are all stored there in structured form; in self-hosted mode this data lives only on your own machine.
- How do I update a self-hosted Enclave to a new version?Run git pull to fetch the latest code, then docker compose up -d to restart and roll forward to the new version. The CHANGELOG flags any breaking changes for each release — just take a quick look before upgrading.
- Can Enclave's medical / legal experts replace a real doctor or lawyer?No, and they shouldn't. Enclave's experts help you organize your thoughts, do your homework, and prepare your questions so you can face real-world professionals better prepared; when there's genuine risk, they'll clearly advise you to seek professional help in person.
- Is Enclave still being updated? Where can I see the progress?Yes, it's being updated. Enclave iterates continuously, and the changes in every release are recorded in the changelog (CHANGELOG) — you can check the latest progress and version history on the site's changelog page.