Tabby
An open-source, self-hosted AI coding assistant with completion, chat, an answer engine and MCP agent mode, free for up to 5 users.
- Free plan
- AI assistant
- Mixed sources
- Verified 13 Nov 2026Sources are dated on this profile
- 4 sourcesLinked, typed and dated
- Vendor and third-party sourcesHow this profile was compiled
- Not an endorsementStack18 has not tested this product
What is Tabby?
Tabby is an open-source (Apache 2.0), self-hosted AI coding assistant from TabbyML, written in Rust. It provides code completion, an answer engine, inline chat, data connectors and an MCP agent mode, running entirely on your own infrastructure with open-weight models.
- Best forTeams that must self-host
- PricingCommunity free (up to 5 users); Team about $19 per user per month; Enterprise custom. Free plan available.
- Main limitationRequires GPU hardware to self-host.
- AlternativesTabnine · Refact.ai · Continue · compare all
Tabby overview: who it is for and what it does.
Tabby is a self-hosted alternative to cloud copilots: install the server on your own GPUs, connect VS Code or JetBrains, and keep code inside your network. It has about 33,000 GitHub stars and partners with Mistral, DeepSeek and Qwen for models.
Community is free for up to 5 users, Team costs about $19 per user a month and Enterprise adds SSO and audit logs. Self-hosting requires GPU hardware, which is the main hidden cost.
Main problem it solves
Gives security-conscious teams AI completion and chat without code leaving their network.
Typical customer segment
Teams of any size that self-host.
Target users
- Security-conscious engineering teams
- Organisations with data-residency rules
- Developers running open-weight models
Primary use cases
- Self-hosted code completion
- Chat and answer engine over internal code
- MCP agent mode
- Air-gapped environments
What makes it distinct
- Fully self-hosted.
- Open source under Apache 2.0.
- Free for up to 5 users.
Tabby features explained.
Assistance
- Completion and chat
- Inline completion and chat.
- Answer engine
- Answers grounded in your code and docs.
Agent and admin
- MCP agent mode
- Agent mode with MCP.
- Enterprise
- SSO and audit logs.
Tabby pricing and plans.
Figures reported consistently by independent sources; vendor page not reviewed directly. Checked 13 Nov 2026 against DEV Community · CostBench · AI Tools Atlas.
Pricing model
Free self-hosted Community edition; per-user Team and Enterprise plans.
Free plan
Yes. Community is free and self-hosted for up to 5 users.
Free trial
Not documented.
Tabby plans and prices (USD)
-
Community
Free
n/a
Self-hosted, up to 5 users.
-
Team
$19 per user per month
Monthly
Reported by trackers.
-
Enterprise
Custom
Custom
SSO/SAML and audit logs.
Prices from 2026 trackers; GPU hardware for self-hosting is an additional cost.
Prices change. Confirm current plans and terms on the vendor's site before you decide.
Visit official website (external) ↗See alternativesTabby integrations, API and platforms.
Important integrations
- VS Code
- JetBrains IDEs
- Vim and Neovim
- Open-weight models (Mistral, DeepSeek, Qwen)
- MCP
- Data connectors
Supported platforms
- Self-hosted server
- IDE extensions
API
Self-hosted server with an API.
Data import and export
Not reviewed for this profile.
Integration availability can depend on plan. Confirm with the vendor before you rely on a specific connector.
How Tabby works as an AI product.
Attributes are listed only where documented. We do not infer autonomy levels, certifications or model capabilities.
Classification: AI assistant. Answers questions and suggests content or code; the user performs the actions. Developer: TabbyML.
| Attribute | Details |
|---|---|
| Supported models | Open-weight models from partners such as Mistral, DeepSeek and Qwen. [4] |
| Tools and integrations | Completion, answer engine, inline chat, data connectors and MCP agent mode. [3] |
| Ability to execute actions | Not documented in the sources reviewed. |
| Execution environment | Not documented in the sources reviewed. |
| Human-in-the-loop controls | Not documented in the sources reviewed. |
| Permissions and access | Not documented in the sources reviewed. |
| Memory and context | Not documented in the sources reviewed. |
| Deployment options | Self-hosted on your own infrastructure. [1] |
| Data handling and privacy | Not documented in the sources reviewed. |
| Security and compliance | SSO/SAML and audit logs on Enterprise. [3] |
| Pricing and usage limits | Not documented in the sources reviewed. |
| Languages and platforms | Not documented in the sources reviewed. |
| Limitations and known constraints | Not documented in the sources reviewed. |
Primary tasks
- Code completion
- Chat over internal code
- Agent tasks via MCP
Best-fit workflows
- Self-hosted environments
- Air-gapped teams
- Open-weight model users
Tabby pros and cons.
Strengths, limitations and best fit. Each point cites a source; editorial summaries are not customer reviews.
Strengths
Limitations
Best for
- Teams that must self-host
- Small teams (up to 5) wanting free AI completion
- Open-weight model users
May be less suited to
- Teams without GPU infrastructure
- Developers wanting autonomous agents
Top Tabby alternatives and competitors.
Tabby alternatives
Teams look for Tabby alternatives when they lack GPU infrastructure, want stronger agents, or prefer a…
Explore →Related productTabnine
A privacy-focused AI coding platform with completions, chat and autonomous agents, deployable as SaaS, in a…
Explore →Related productRefact.ai
An open-source AI coding agent and autocomplete for VS Code and JetBrains that can be self-hosted and…
Explore →Related productContinue
Open-source AI coding extensions for VS Code and JetBrains, a CLI and cloud agents that run…
Explore →Tabby compared with similar AI coding agents.
Summaries come from each profile. This is a side-by-side aid, not a ranking.
| Product | Interface | Model choice | Free tier | Pricing model | Human oversight |
|---|---|---|---|---|---|
| Tabby | Self-hosted server with IDE extensions | Open-weight models you host | Yes (up to 5 users) | Free self-hosted; per-user Team | Not documented in sources reviewed |
| Augment Code | VS Code, JetBrains, CLI and Slack | Frontier models; credit cost varies | No (trial reported) | Credit-based subscriptions | Not documented in sources reviewed |
| Blackbox AI | VS Code agent, IDE, web and CLI | Several model vendors | Unclear after late-2026 change | Enterprise committed spend (earlier $10 to $40) | Not documented in sources reviewed |
| Cline | VS Code extension and CLI | Many providers, bring your own key | Yes (open source) | Free software; pay model usage | Per-action approval and checkpoints (third-party) |
| Continue | VS Code and JetBrains extensions, CLI, cloud agents | Any model, including local | Yes (open source) | Free software; PAYG tokens or Team seats | Checks reported as GitHub status checks |
| GitHub Copilot | IDE extensions, CLI, desktop app and GitHub.com | Several vendors' models, plus third-party agents | Yes (Copilot Free) | Subscription with monthly AI credits | Approvals, rollback and credit limits documented |
Tabby vs competitors: head-to-head comparisons.
Tabby discussions and buying questions.
Tabby questions and answers
What is Tabby used for?
Tabby is an open-source (Apache 2.0), self-hosted AI coding assistant from TabbyML, written in Rust. It provides code completion, an answer engine, inline chat, data connectors and an MCP agent mode, running entirely on your own infrastructure with open-weight models.
Is Tabby free?
Yes. Community is free and self-hosted for up to 5 users.
How much does Tabby cost?
Community free (up to 5 users); Team about $19 per user per month; Enterprise custom.
Which models does Tabby support?
Open-weight models from partners such as Mistral, DeepSeek and Qwen.
What are the limitations of Tabby?
Requires GPU hardware to self-host. Agent capabilities are lighter than dedicated agents. Prices are tracker-reported.
Which platforms does Tabby run on?
Tabby is documented as available on: Self-hosted server, IDE extensions.
Is Tabby an AI agent or an AI assistant?
We classify Tabby as: AI assistant. Answers questions and suggests content or code; the user performs the actions. The classification is based on vendor documentation and is not an endorsement.
What are the main alternatives to Tabby?
The main alternatives to Tabby covered on Stack18 are Tabnine, Refact.ai, Continue and GitHub Copilot. See the Tabby alternatives guide for a side-by-side table.
When was the Tabby profile last verified?
Sources for this profile were last reviewed on 13 Nov 2026. 4 sources are listed under Sources, and prices and plans can change, so confirm with the vendor.
Where this information comes from.
- Tabby Review 2026 DEV Community Third-party Accessed 13 Nov 2026
- Tabby pricing CostBench Third-party Accessed 13 Nov 2026
- Tabby ML pricing AI Tools Atlas Third-party Accessed 13 Nov 2026
- Tabby EveryDev Third-party Accessed 13 Nov 2026
Further official pages: Tabby pricing (external)
Plans, features and status were checked against Tabby's published pages and the third-party sources listed. Agent behaviour was not tested, and security certifications were not independently verified.
Stack18 has not tested this product or received anything from the vendor for this profile. Editorial summaries of vendor information are not endorsements. Confirm details with the vendor before buying.
Evaluating Tabby?
Check the vendor's current pricing and terms, compare it with similar products, or read how others approach the decision. If off-the-shelf is not the answer, Stack18 builds around how your business works.
Logos provided by Logo.dev. Product names and logos are trademarks of their respective owners and appear for identification only; they do not imply endorsement.