AI Frameworks · SillyTavern

SillyTavern

SillyTavern is a locally-installed web interface for interacting with multiple LLM APIs (OpenAI, Claude, Mistral, Horde, NovelAI, etc.), image generation engines, and TTS models. It runs on NodeJS with minimal hardware requirements and offers a feature-rich UI including Visual Novel Mode, character lorebooks, and third-party extension support.

Talk to DEV.co Discuss Implementation

Source: GitHub — github.com/SillyTavern/SillyTavern

30.4k

GitHub stars

5.7k

Forks

JavaScript

Primary language

AGPL-3.0

License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

Field	Value
Repository	SillyTavern/SillyTavern
Owner	SillyTavern
Primary language	JavaScript
License	AGPL-3.0 — OSI-approved
Stars	30.4k
Forks	5.7k
Open issues	522
Latest release	1.18.0 (2026-05-03)
Last updated	2026-07-07
Source	https://github.com/SillyTavern/SillyTavern

What SillyTavern is

JavaScript-based frontend supporting multiple LLM backends and APIs via unified interface, with optional local inference capability (GPU-accelerated on NVIDIA), Docker deployment, mobile-responsive layout, and plugin/extension architecture. Licensed under AGPL-3.0 with ~300 contributors and 3+ years of active development since forking from TavernAI.

Quickstart

Get the SillyTavern source

Clone the repository and explore it locally.

terminalbash

1git clone https://github.com/SillyTavern/SillyTavern.git2cd SillyTavern3# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

AI hobby enthusiasts with multi-backend support needs

Users wanting to experiment across OpenAI, Claude, local models, and niche APIs benefit from the unified interface without vendor lock-in. Extensive prompt engineering options and WorldInfo feature suit power users.

Local-first LLM inference with UI customization

Teams running private or on-prem models via KoboldAI/Ollama backends can leverage ST's feature-complete frontend without external service dependencies. Customizable UI and character system reduce development friction.

Creative writing and character-driven AI applications

Visual Novel Mode, character lorebooks, and prompt templates are tailored for fiction, roleplay, and narrative content creation. Multi-language support (8 documented translations) expands reach.

Implementation considerations

NodeJS 20+ is hard requirement; local GPU inference requires NVIDIA 3000-series+ (min 6GB VRAM) for practical model hosting. CPU-only inference is slow.
AGPL-3.0 requires any network-accessible modifications to have source code available to users. Internal use does not trigger this; hosted/SaaS wrappers do.
Minimal dependencies described in README; full dependency audit and supply-chain assessment needed before production deployment.
No built-in authentication/authorization—proxy layer or reverse-proxy (nginx/HAProxy) with auth required for multi-user or external access.
522 open issues and active development suggest rapid iteration; plan for compatibility testing across minor releases and document pin points.

When to avoid it — and what to weigh

Requiring commercial closed-source derivative rights — AGPL-3.0 mandates source disclosure for network-accessible modifications. Any SaaS wrapper or hosted commercial service requires releasing modified code. Internal enterprise use is permitted.
Needing enterprise SLA and vendor support — SillyTavern is community-maintained passion project with no paid support tiers, SLA guarantees, or official enterprise contracts. Support via Discord and GitHub issues only.
Minimal infrastructure or offline-only deployment — Requires NodeJS 20+ runtime and internet access for most LLM backends (except local inference). Not suitable for air-gapped or extremely resource-constrained environments.
Strict liability and warranty requirements — License explicitly disclaims warranty and liability. Not suitable for regulated industries (healthcare, finance) without extensive custom compliance work and legal review.

License & commercial use

GNU Affero General Public License v3.0 (AGPL-3.0). This is a copyleft license requiring source disclosure for any derivative works, especially network-accessible modifications. Internal use is permitted; hosted/SaaS derivatives require source code availability to end users.

Internal commercial use (e.g., as employee tool) is permitted under AGPL-3.0. However, any SaaS, hosted service, or web-accessible modification must disclose source code. Selling or licensing compiled/packaged variants without source disclosure violates terms. Consult legal counsel before commercial deployment. No commercial support or warranty provided.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

Signal	Assessment
Maintenance	Active
Documentation	Strong
License clarity	Clear
Deployment complexity	Low
DEV.co fit	Good
Assessment confidence	High

Security considerations

AGPL-3.0 license includes no warranty or liability. NodeJS runtime and third-party extension ecosystem introduce supply-chain risk (audit dependencies and extensions). Network-facing deployment requires authentication/authorization layer (not built-in). API credential handling (OpenAI, Claude keys, etc.) must follow secrets management best practices. No independent security audit data provided; user-run code on local/private infrastructure only.

Alternatives to consider

OpenAI ChatGPT / Claude.ai web interface

Hosted, officially-supported, no infrastructure required. Trade-off: vendor lock-in, closed-source, no local inference, no customization.

Ollama (local LLM runtime) + custom UI

Lightweight local model hosting; requires custom frontend development. Trade-off: no unified multi-backend support, more engineering overhead.

LocalAI / LM Studio

Focused on local model inference with built-in UI. Trade-off: limited to local/open models, no multi-cloud backend support, smaller community.

Software development agency

Build on SillyTavern with DEV.co software developers

Our team can guide you through setup, AGPL-3.0 compliance, backend integration, security hardening, and custom extensions. Contact us for a no-obligation consultation.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

skyvern llm-ui llmware

Related on DEV.co

Explore the category and the services that help you build with it.

AI Frameworks AI application development API development Custom software development Talk to DEV.co

SillyTavern FAQ

Can I use SillyTavern in a commercial SaaS product?

No, not without releasing source code. AGPL-3.0 requires any network-accessible derivative to provide source to users. Internal commercial use (employee tool, no external users) is permitted.

What are minimum hardware specs?

NodeJS 20+ only requirement. For local model inference, NVIDIA 3000-series GPU with 6GB+ VRAM recommended. CPU-only inference works but is slow.

Is there official enterprise support?

No. Support is community-driven via Discord and GitHub. No SLA, paid support tiers, or vendor contracts available.

What LLM backends are supported?

OpenAI, Claude, Mistral, OpenRouter, Horde, KoboldAI/CPP, TabbyAPI, NovelAI, Ooba, and more. See documentation for full list and configuration details.

Custom software development services

DEV.co helps companies turn open-source tools like SillyTavern into production software. Our software development services cover the full lifecycle — architecture, web development, integration, and maintenance — delivered by software developers and web developers who ship. Engage our software development agency to implement or customize it for your ai frameworks stack.

Need help integrating or deploying SillyTavern?

Our team can guide you through setup, AGPL-3.0 compliance, backend integration, security hardening, and custom extensions. Contact us for a no-obligation consultation.

Talk to DEV.co Schedule a Consultation