AiBOX.im

SillyTavern

Roleplay AI chat frontend that runs locally in your browser

— views · — installs
WindowsmacOS Difficulty ●●●○○ stable v1.19.0
SillyTavern icon

SillyTavern · icon from the registry

System requirements

Node.js versions

Recommended
Node >=18
Supported
Node >=18 <25
Not supported
16.x

Yours missing or too old? AiBOX installs a supported Node.js next to it and leaves the one you already have untouched.

Platforms

WindowsmacOS

Roleplay AI chat frontend that runs locally in your browser

Registry data updated: 2026-11-15

Config & data

Config file written to

~/SillyTavern/data/default-user/settings.json

  • ~/AppData/Roaming/SillyTavern
  • ~/Library/Application Support/SillyTavern

Everything stays on your own machine — nothing is uploaded. To uninstall, delete these folders.

Install in three steps

  1. 01

    Install AiBOX

    One portable file for Windows (macOS coming). Double-click and it runs — no installer, no administrator rights.

    Go to download →
  2. 02

    Pick SillyTavern

    Open the AiBOX app catalog, find SillyTavern and press Install. AiBOX checks what your machine already has and installs only what is missing.

  3. 03

    Follow the wizard

    It downloads with a checksum check, writes the config and fills in your API key if the app needs one. You just click.

    ✓ Installed until it works.

Prefer to do it by hand? The full written guide is right below. Read the guide ↓

Install guide

The same steps in written form — follow along at your own pace.

One line: SillyTavern is a roleplay-focused AI chat frontend that runs on your own computer — you use it in a browser, connect practically any LLM (cloud API or a local model), and every character card, preset and chat stays on your machine.

What you get

  • A local web chatroom at 127.0.0.1:8000: character cards, group chats, lorebooks, memory tools
  • Model freedom: OpenAI, Claude, Google, OpenRouter, DeepSeek, free crowd-sourced AI Horde, KoboldCpp, Ollama, and more
  • An extension ecosystem — translation, image gen, voice, TTS, expressions, retrieval — installable from a panel

Before you start (5 minutes)

NeedNotes
Windows 10/11 or macOS 12+Linux works too; no admin rights, everything lives in your home folder
Node.js 18+LTS from nodejs.org; check with node --version
GitRequired for extensions and updates (git-scm.com on Windows)
A model sourceA cloud API key, or local Ollama / KoboldCpp. Zero-cost trial: AI Horde with key 0000000000

Install

Route A: install with AiBOX (recommended)

  1. Open AiBOX → app directory → SillyTavern → Install
  2. AiBOX clones the official source into ~/SillyTavern (hash-verified, CN traffic auto-routes through a proxy channel) and runs start.sh to fetch dependencies
  3. When it finishes, open http://127.0.0.1:8000 in your browser

Route B: manual install (official commands)

git clone --depth 1 -b 1.19.0 https://github.com/SillyTavern/SillyTavern.git ~/SillyTavern
cd ~/SillyTavern && ./start.sh        # Windows: double-click Start.bat, or run in Git Bash

start.sh always runs npm install (idempotent) before starting the server — download cut off halfway? Just rerun the same command and it picks up where it left off.

Versions & updating (release vs staging)

  • Upstream keeps two branches: release (recommended, cut roughly monthly, stable) and staging (multiple daily updates, can break any moment — enthusiasts only)
  • AiBOX pins a release tag (e.g. 1.19.0). To upgrade: run git fetch && git checkout 1.x.y (or git pull) inside ~/SillyTavern, then ./start.sh — dependencies refresh automatically
  • Want the newest features: git checkout staging && ./start.sh; roll back the same way with git checkout release

Configuration & verification

Connect a model (which API type to pick)

Top bar → API Connections → pick the type in the first dropdown, then the source:

TypeUse it whenTypical sources
Chat CompletionCloud conversational APIs (easiest)OpenAI, Claude, Google, OpenRouter, DeepSeek
Text CompletionLocal / continuation-style models — needs an Instruct templateOllama, KoboldCpp, TabbyAPI, llama.cpp
AI HordeCompletely free volunteer compute, variable wait timeskey 0000000000 works out of the box
OthersNiche needsNovelAI, Mancer, Pollinations, …
  • Click Connect — the status must read Valid before you chat
  • For Text Completion, set the model's Instruct template on the Advanced Formatting page (usually stated on the model's download page)
  • Settings live in ~/SillyTavern/data/default-user/ — copy that folder to migrate

Importing character cards

  1. Top bar → Character Management → Import: drop a PNG character card, paste a card URL, or pick from the built-in list
  2. Web-hosted imports must point at a domain listed in config.yaml under whitelistImportDomains (a guard against pulling arbitrary remote content)
  3. Select the character, send a message — an in-persona reply means the import worked

Presets & sampling, briefly

  • User Settings → AI Response Formation: the preset dropdown (Classic, Alpaca, …) decides how the prompt is assembled; beginners should stay on the default
  • Temperature, Top P, repetition penalty are the three knobs worth touching: reply too stiff → raise temperature (0.9–1.1); rambling nonsense → lower it (0.6–0.8)
  • After switching models, load a preset recommended for that model and double-check the Instruct template under Advanced Formatting

What "success" looks like

  1. http://127.0.0.1:8000 opens (trust the URL printed in the terminal)
  2. API Connections shows Valid
  3. Pick any built-in character, send one line, get an in-persona reply within seconds

FAQ

  • npm could not be found: install Node.js 18+ and open a fresh terminal before rerunning ./start.sh
  • Port already in use: ./start.sh --port 8001 (arguments are passed straight through to the server; --listen exposes it on your LAN, --browserLaunchEnabled false skips auto-opening the browser)
  • Install failed halfway: rerun ./start.sh to resume — finished node_modules work is not redone
  • Extensions won't install: git is required; use Extensions → Download Extensions & Assets → Load Asset List → Download
  • API won't connect: verify the key and account credit; for local models make sure the server is running (ollama serve) and the port matches
  • Access from phone / LAN: start with --listen and configure a whitelist (./start.sh --listen --whitelist) — never expose it to the open internet unprotected

Uninstall

rm -rf ~/SillyTavern                                  # program + character cards + chats live here — back up data/ first
rm -rf ~/AppData/Roaming/SillyTavern                  # Windows global-mode leftovers
rm -rf ~/Library/Application\ Support/SillyTavern     # macOS global-mode leftovers

AiBOX's Manage → Uninstall does this for you. Export your characters and chats before deleting.

More: official docs · quick start · GitHub

Model / API key setup

AiBOX fills this in for you during the wizard — this section is for setting it up by hand.

OpenRouter

You need an API key from the provider.

~/SillyTavern/data/default-user/settings.json

Get a key ↗

Anthropic (Claude)

You need an API key from the provider.

~/SillyTavern/data/default-user/settings.json

Get a key ↗

OpenAI

You need an API key from the provider.

~/SillyTavern/data/default-user/settings.json

Get a key ↗

Google AI Studio

You need an API key from the provider.

~/SillyTavern/data/default-user/settings.json

Get a key ↗

Claude subscription

Sign in with your account instead — no key to copy.

~/SillyTavern/data/default-user/settings.json

Get a key ↗

Ollama (local)

Local model — no API key needed.

~/SillyTavern/data/default-user/settings.json

Common questions

Which platforms does SillyTavern run on?

Runs on Windows / macOS — install it with AiBOX on any of them.

How hard is SillyTavern to install?

Difficulty 3 out of 5. AiBOX does the download and setup; your part is clicking through the wizard and pasting a key if the app needs one.

Where does SillyTavern keep my data?

Everything stays on your own machine: ~/AppData/Roaming/SillyTavern / ~/Library/Application Support/SillyTavern. To uninstall, delete these — nothing is ever uploaded.

Something not working?

An outdated step, an error you do not understand, a typo — tell us and we will fix this page.

Report a problem →

Version & updates

Current version
v1.19.0
Registry data updated
2026-11-15

Version numbers are re-checked by our daily probe — when upstream ships a new release, this page follows.

Questions or problems?

A wrong step, a confusing error, a typo — open an issue and someone will look.