Stuck on something? Drop into the Discord - fastest way to get an answer.

Does it run on Windows?

Yes - natively on Windows 10 1809+ and Windows 11 (x64). Portable ZIP or NSIS setup wizard from the latest release. Secrets are DPAPI-encrypted. WSL also works if you prefer the Linux path. See installation for limits (no embedded floating terminal, no Kitty image preview).

Do I need Neovim?

No. Neovim is an optional editor panel for users who want one. Empryo works fully from the chat alone. If you do want the panel, Empryo can auto-install Neovim and a Nerd Font on first launch.

Is there a GUI, or is it terminal-only?

Both. The desktop app is the full workspace in a native window — multi-tab agents, the live genome map, embedded editor and terminals — and the terminal build carries the TUI plus a scriptable --headless mode for CI. Same engine, same Genome, same sessions under all three.

Are the benchmark claims real?

They're the most auditable numbers in the category: each agent runs on its own API key, so the provider's bill — not anyone's telemetry — is the judge, and the grading is hidden acceptance tests on real bugs with fixes merged past the models' training cutoffs. The harness, tasks and raw results are public; reruns are two keys and one command. Full write-ups on the benchmarks page.

Can I import my Claude Code, Codex, or OpenCode sessions?

Yes — all three. /import (or the desktop wizard's Import step) scans ~/.claude, ~/.codex, and OpenCode's local database, shows what it found, and migrates transcripts into Empryo's session store — each session imported individually, so one odd file never sinks the batch.

Do I need an API key?

You don't even need to copy one. Three paths, pick your favorite:

Browser login (fastest). Type /login inside Empryo — it opens LLM Gateway in your browser, which mints a key and hands it straight back over a local loopback. One login unlocks 200+ models across 40+ providers. Pay with account credits (pay-as-you-go, /login) or a DevPass subscription (flat plan, billed per request — /login llmgateway-devpass). In the desktop app both live under Settings → Keys.

Bring a key you already have. Any individual provider works:

empryo --set-key anthropic sk-ant-...
empryo --set-key llmgateway sk-...   # one key, every model

Local, free. Install Ollama or LM Studio. Empryo auto-detects both. Cost tracking reports $0.00 for local runs.

brew install ollama && ollama pull llama3.3

How much does it cost to run?

Cloud costs depend on your model. Empryo shows the live USD number as you work, broken down per model and per subagent. Cache-aware pricing for all major providers.

Rough ballpark on a mid-size repo: a one-file refactor under $0.05 on Sonnet with caching, a cross-file rename often under $0.05 because the compound tool runs in three steps, an all-day session with a mix of Haiku exploration and Sonnet coding typically stays under $5.

Local models are free. Use the task router to push exploration to a cheap model and reserve the strong model for code.

Which model should I pick?

Claude Sonnet 4.6 for balanced code quality — the everyday default.
Claude Haiku 4.5 for fast exploration and cleanup. In our benchmarks Haiku + the Genome fixed more real bugs than a rival agent at half the wall clock — cheap models go further with a code graph underneath.
Claude Opus 4.8 when the bug is genuinely hard (it was the only config that cracked our benchmark's deepest task on the first try).
Local Llama/Qwen/DeepSeek if you want $0 and privacy.

Mix them via /router. See the task router recipe.

Does it work without internet?

With local LLMs (Ollama, LM Studio), yes - fully offline for model calls. The Genome, LSP, tree-sitter, and everything else is local anyway.

How does Empryo edit code?

Four tools, picked by the one that fits:

AST editing for TypeScript and JavaScript. Addresses functions, classes, methods by name - not text. A return-type change costs a handful of tokens.
Multi-edit for batched text edits on the same file, applied atomically with line-offset tracking.
Edit for single text edits with line-anchored matching on any file (JSON, YAML, Markdown, Python, Rust, Go, etc).
LSP rename and refactor for workspace-wide symbol renames, move symbol, extract function, organize imports. Compiler-verified, every file updated.

The agent picks automatically.

Is my code uploaded anywhere?

Only what you send to your chosen LLM provider, same as any other AI coding tool. Empryo does not upload anything to our servers. There is no Empryo backend.

Privacy patterns (/privacy add) block sensitive files from the agent entirely. Built-in blocks already cover .env, .pem, credentials, id_rsa, .npmrc, .netrc, and more.

Can I use it for work?

Yes. Empryo is free to use — personal projects, startups, enterprise, client work, whatever. You bring your own model key (or run local models), so the only thing you ever pay for is the tokens you choose to spend with your provider.

What languages does code intelligence cover?

Empryo ships tree-sitter grammars for 30 languages, so it parses structure — symbols, imports, outlines — in: TypeScript, JavaScript, Python, Go, Rust, Java, C, C++, C#, Ruby, PHP, Swift, Kotlin, Scala, Lua, Elixir, Dart, Zig, Bash, OCaml, Objective-C, CSS, HTML, JSON, TOML, Vue, ReScript, Solidity, TLA+, and Emacs Lisp. Other file types (Markdown, YAML, SQL, GraphQL, Dockerfile, Terraform, and more) are detected for navigation but not deeply parsed.

LSP coverage depends on what's installed via Mason - typical set: TypeScript, JavaScript, Python, Go, Rust, Java, Kotlin, Scala, C#, Swift, Dart, Elixir, OCaml, Lua, C, C++, Ruby, PHP, Zig, Bash, CSS, HTML, JSON, TOML, YAML, Dockerfile, Vue.

AST editing (ts-morph) is TypeScript and JavaScript only. Everything else uses LSP and tree-sitter.

Can I install LSP servers from Empryo?

Yes. /lsp install opens a browser over Mason's registry (576+ packages). Empryo installs into ~/.empryo/lsp-servers/ via npm, pip, cargo, go, or GitHub releases depending on the package.

How do sessions work?

Every conversation is auto-saved as JSONL under ~/.empryo/sessions/. Resume with empryo --headless --session <prefix> or press Ctrl+P in the TUI. Each session carries its checkpoint git tags - undo all the way back to any earlier turn.

Can I run this over SSH?

Yes. The TUI is pure terminal output (OpenTUI with Kitty graphics for inline images). SSH, tmux, and remote shells all work.

What about multi-tab work?

Up to 10 tabs per project. Each has its own session, model, mode, and checkpoint history. Tabs are aware of each other: when one claims a file, the others see the claim and skip it. Git operations hard-block during cross-tab dispatch so you never commit partial changes. See tabs.

How do themes work?

39 built-in themes (Catppuccin, Dracula, Gruvbox, Tokyo Night, Rose Pine, Kanagawa, Nightfox, Cyberdream, Oxocarbon, Ember, Vesper, and more). /theme opens a live-preview picker. Custom themes go in ~/.empryo/themes/<name>.json and hot-reload on save. See themes.

How do I get help?

Troubleshooting - common issues.
/diagnose inside Empryo - LSP, tree-sitter, semantic indexing health check.
/status - full system dashboard.
GitHub issues.