English · Русский
Iva is a self-hosted Telegram AI assistant with layered memory that turns your messages into an Obsidian-compatible vault. You talk, it files: voice notes, photos, forwarded posts and decisions become plain-markdown cards it actually remembers. Everything runs on your own server, with your keys and your data. One command installs it:
curl -fsSL https://raw.githubusercontent.com/smixs/iva/main/install.sh | bash- 🎙️ Voice — voice, audio and video notes transcribed with Deepgram nova-3; auto-detects ru/uz/en.
- 👁️ Vision — photos described by your provider's own vision model; no extra key, no extra bill.
- 🧠 Layered memory — remembers across months, not just the current chat window.
- 📇 Personal CRM — who your people are, what you agreed, when to follow up.
- 🔎 Search by meaning — BM25 plus link-graph rerank, any language; optional vector mode with one key.
- 🧭 Decision cards — what you chose, when and why; old versions stay in a dated History.
- ⏰ Tasks & reminders — priorities, due dates and a morning digest.
- 🌐 Web search — four pluggable providers: Tavily, Exa, Parallel or Brave.
- 🧩 Skills & MCP — drop one file to add a procedure or connect an MCP server; keys stay in
.env. - 🛡️ Safe to forward — links, PDFs and other people's messages are screened before the model reads them.
- 📊 Token accounting — every model step is logged;
/usagereports it for free.
- Every message lands verbatim in a daily markdown log — nothing is paraphrased on arrival.
- A nightly rollup at 04:00 distills day → week → month → year into schema-validated cards; facts that change get rewritten, not piled up.
- One core file,
CORE.md(≤1,200 chars), rides in every prompt — Iva knows you before it searches anything.
Full architecture and search internals: docs/memory.md.
- Get a bot token from @BotFather.
- Run the one-line installer above on any Ubuntu/Debian box — a fresh VPS or your own machine.
- Message your bot. The wizard picks your Telegram ID out of that message, finishes setup, and Iva confirms right in the chat that it's live.
Headless installs take --skip-setup or --non-interactive. Wizard walkthrough and an SSH primer for first-time VPS owners: docs/install.md.
The bridge long-polls Telegram, so no public HTTPS, domain or webhook is needed. The agent, the bridge and five memory timers run as systemd user units on your box — operations live in docs/deploy.md.
| What | Cost |
|---|---|
| OpenCode Zen — model API | ~$5/mo |
| Ollama Cloud — model API | ~$20/mo |
| VPS — 512 MB is enough | $4–5/mo |
Pick one provider; the default model is deepseek-v4-pro with a 131k context. About $9/mo total, no markup. Voice rides Deepgram's free starter credit. Model lists, limits and the search-provider matrix: docs/providers.md.
Inbound content passes a prompt-injection sanitizer, every reply passes a secret-redaction gate, and the user allowlist fails closed — an empty list answers nobody. Your memory is a private git repo you own; the honest boundary is that the model and transcription are cloud APIs you choose and pay for. Gate internals: docs/security.md.
| In Telegram | On the server |
|---|---|
/help · /task · /digest · /new · /usage |
iva status · iva update · iva doctor · iva logs |
Full reference, including /usage breakdowns by model and by source: docs/cli.md.
Install · Configuration · Memory · Providers · Security · Deploy · CLI · Extending · FAQ · Troubleshooting
Документация на русском → docs/ru/
eve, Vercel's agent framework, runs the agent; Node 24's built-in SQLite runs the search index — no separate database. Iva grew out of agent-second-brain and autograph — that story is in docs/memory.md.
MIT — take it, change it, run it on a hundred servers; just don't blame anyone if something breaks.