Tabby is an open-source AI coding assistant designed to enhance developer productivity with features like code completion, an answer engine for instant coding questions, and inline chat, all integrated directly into your IDE.

Can Tabby be self-hosted?

Yes, Tabby offers flexible deployment options, including self-hosting on your own infrastructure. This allows for greater control over data privacy and security, and supports consumer-grade GPUs.

What IDEs does Tabby support?

Tabby supports a wide range of popular IDEs and editors, including VS Code, Neovim, IntelliJ, Eclipse, Android Studio, AppCode, CLion, GoLand, PhpStorm, PyCharm, Rider, RubyMine, and WebStorm.

What AI models does Tabby use?

Tabby is compatible with various coding LLMs, including StarCoder, CodeLlama, DeepSeekCoder, CodeGemma, CodeQwen, Qwen2.5-Coder, Codestral, Mistral, and Yi-Coder, allowing users to choose based on their needs and licensing.

Is Tabby free to use?

Yes, Tabby offers a free community plan and is open-source, allowing developers to use and contribute to the project. Team and Enterprise plans with additional features are also available.

Launching soon — get early access:

/

AI Tools

/

Tabby

Open Source

Open-source coding assistant

Self-hostable AI

Best for privacy

Highly configurable

IDE-integrated AI

Secure, flexible, and transparent AI coding assistant for every developer.

Last updated: Jun 1, 2026

OPEN_SOURCE

Visit Site GitHub

33.0k

Try Demo

tabby.tabbyml.com

Visit tabby.tabbyml.com

Quick Answer

Tabby is an open-source, self-hostable AI coding assistant that integrates with popular IDEs to provide code completion, instant answers, and inline chat. It offers developers control over their data and deployment, supporting both cloud and on-premises setups. Tabby is ideal for individuals and teams prioritizing privacy, security, and customization in their AI-powered development tools.

About Tabby

Tabby is an open-source AI coding assistant that integrates directly into your IDE, providing real-time code completion, instant answers, and inline chat. It offers flexible deployment options, including self-hosting and cloud, ensuring privacy and control over your development workflow. Tabby supports a wide range of programming languages and IDEs, powered by various open and proprietary models.

What Makes It Unique

Tabby stands out as an open-source, self-hostable AI coding assistant that prioritizes transparency, security, and configurability, offering a powerful alternative to proprietary solutions with flexible deployment options.

AI-extracted from public sources. May contain errors. ·

Methodology

Are you Tabby? Claim listing

Use Cases & Target Users

Use Cases

Coding

Workflow Automation

Best For

Developers seeking privacy and control over their AI coding assistant

Teams requiring self-hosted AI solutions for security compliance

Engineers looking for highly configurable coding AI

Developers working with diverse programming languages and IDEs

Organizations wanting transparent AI tools for their development workflow

Company Size

SOLO

SMALL

MEDIUM

ENTERPRISE

Fit Score

Complexity

Intermediate

Target Team Size

INDIVIDUAL

SMALL_TEAM

LARGE_TEAM

Target Skill Level

INTERMEDIATE

EXPERT

Key Features

Code completion with AI suggestions

Instant coding answers (Answer Engine)

Real-time AI chat directly in your code (Inline Chat)

Data connectors for enriched AI assistance

Flexible deployment (Cloud & Self-hosting)

Software supply chain safety (Open-source)

Technical Specifications

API Available

SDK Available

MCP Compatible

Open Source

NOASSERTION

Team Features

SSO/Enterprise

Webhook Support

Has Demo

Function Calling

Browser Extension

Sandbox Mode

Audit Logs

Free Plan

No Signup

Works Offline

Requires Host App

Framework

MCP Server

IDE Plugin

Inference API

SCIM Provisioning

Developer Access

API

ENTERPRISE

Deployment:

HYBRID

Uses models:15

Pros

+Open-source and transparent, enhancing trust and security
+Flexible deployment options (cloud and on-premises/self-hosted)
+Highly configurable to fit existing workflows and infrastructure
+Supports a wide range of IDEs and programming languages
+Offers advanced AI capabilities like code completion, inline chat, and answer engine
+No reliance on external DBMS or cloud services for core functionality

Limitations

-Requires technical expertise for self-hosting and configuration
-Performance can be dependent on local hardware (GPUs)
-Some advanced models may have non-production licenses or require commercial licenses
-No explicit mention of advanced team collaboration features beyond user/group management
-Lack of specific data export formats mentioned for easy migration

Powered By

Base Models

CodeLlamaStarCoderCodeGenQwen2.5-Coder-1.5BStarCoder2-3BQwen2.5-Coder-7BDeepSeekCoder-6.7B

Uses Models

StarCoder-1BStarCoder-3BStarCoder-7BStarCoder2-3B