Key Takeaways
- Chatbox (source at github.com/chatboxai/chatbox) has a GPLv3-licensed open-source core with 41,600+ GitHub stars, built by its own team.
- It installs as a native app on Windows, macOS, Linux, iOS, and Android, and also runs with no install at web.chatboxai.app.
- The free app connects to your own API keys (OpenAI, Anthropic Claude, Google Gemini) or a local server such as Ollama or LM Studio, with conversations stored on your device by default.
- Chatbox AI, a separate paid hosted service from the same company, sells API access, web search, image chat, and document chat through Lite, Pro, and Pro+ plans starting at $3.99/month.
- It supports image generation via DALL-E-3 when connected to an OpenAI API key, alongside Markdown, LaTeX, and code-syntax rendering, and is the only one of five compared chat-client apps with a confirmed affiliate program.
π In One Sentence
Chatbox is a GPLv3-licensed AI chat client with 41,600+ GitHub stars, distributed as native apps for Windows, macOS, Linux, iOS, and Android plus a no-install web version, free to use with your own API keys or a local Ollama/LM Studio server, with an optional paid Chatbox AI hosted service on top.
π¬ In Plain Terms
Chatbox isn't a server you deploy -- it's an app you download and install like any other desktop or mobile app, or open in a browser tab with no install at all, and you either type in your own API key or pay the company behind it for access instead.
πNote: This review is the deep-dive companion to Chatbox's entry in the Local LLM Software Directory -- see that page for how Chatbox compares at a glance to dozens of other local AI tools.
Quick Facts: Chatbox
- License: GPLv3 (core client) -- open source with copyleft terms; redistributing a modified version requires releasing it under GPLv3 as well
- Platforms: Native apps for Windows, macOS, and Linux; mobile apps for iOS and Android; web version at web.chatboxai.app with no install
- GitHub: github.com/chatboxai/chatbox -- 41,600+ stars as of September 2026
- Official site: chatboxai.app
- Maintainer: The Chatbox AI company team, not an individual developer
- Cost: Free app with your own API keys or a local server; optional Chatbox AI hosted plans from $3.99/month
- Core function: Cross-platform AI chat client for cloud model providers and local model servers
- Model providers supported: OpenAI, Azure OpenAI, Anthropic Claude, Google Gemini, Ollama, LM Studio, and other OpenAI-compatible endpoints
What Is Chatbox?
Chatbox is an AI chat client with a GPLv3-licensed open-source core, distributed as installable apps rather than a self-hosted server you deploy. You download a native installer for Windows, macOS, or Linux, install a mobile app on iOS or Android, or open web.chatboxai.app in a browser with no install at all -- the same account and interface work across all of them.
It does not run models itself. Instead, it is a front end that talks to whichever backend you connect: cloud providers such as OpenAI, Anthropic Claude, Google Gemini, and Azure OpenAI using your own API keys, or a local server such as Ollama or LM Studio using its OpenAI-compatible API endpoint.
What sets Chatbox apart from most other open-source chat clients is that the company behind it also sells a hosted service, Chatbox AI, on top of the free app. Instead of entering your own API keys, a paid Chatbox AI plan gives you API access, web search, image chat, and (on higher tiers) document chat and MCP integration directly through the same client.
History and Development
Chatbox is built and maintained by the team behind the Chatbox AI product, rather than a single independent maintainer, and the project is distributed both as a free, open-source client and as a paid hosted service layered on top of it.
The GitHub repository's feature history shows a client focused on cross-platform reach from early on -- native builds for desktop and mobile rather than a single web deployment, multi-provider support for OpenAI, Anthropic Claude, Google Gemini, and Azure OpenAI, and local-model connectivity through Ollama-compatible endpoints, alongside Markdown, LaTeX, and code-syntax rendering, a prompt library, and image generation via DALL-E-3.
A significant point for anyone evaluating Chatbox commercially is the license. The core client ships under GPLv3, a copyleft license: unlike a plain MIT license, redistributing a modified version of Chatbox requires releasing your changes under GPLv3 as well. This is a stricter condition than the MIT license used by Page Assist, Big-AGI, and NextChat, and worth checking before building a commercial fork.
What Can You Actually Do With Chatbox?
- Chat with a cloud model or a local model in the same app. Enter your own API key for OpenAI, Anthropic Claude, Google Gemini, or Azure OpenAI, or point Chatbox at a local Ollama or LM Studio server using its API base URL.
- Use the same client on desktop, mobile, and web. Native installers for Windows, macOS, and Linux, native apps for iOS and Android, and a no-install web version at web.chatboxai.app all share the same interface.
- Generate images with DALL-E-3. When connected with an OpenAI API key, Chatbox can send image-generation requests directly from a chat.
- Keep conversations on your device by default. The free app stores data locally rather than uploading it to a Chatbox-operated server, according to the project's GitHub documentation.
- Read Markdown, LaTeX, and syntax-highlighted code. Model responses render formatted text, math notation, and code blocks rather than plain unformatted output.
- Reuse prompts from a built-in prompt library, and quote earlier messages when composing a new one, instead of retyping context by hand.
- Pay for API access instead of managing your own keys. The paid Chatbox AI plans add web search, image chat, and (on Pro and Pro+) document chat and MCP integration on top of the free app's feature set.
- Use the interface in one of 9 supported languages, set from within the app's settings.
Get Chatbox: Desktop, Web, and Mobile
Chatbox is a genuine downloadable end-user app across six surfaces, not a self-hosted server -- get it directly from the official site, GitHub releases, or the web version below:
Platform | Get It |
|---|---|
| Windows / macOS / Linux | Download installers from chatboxai.app |
| Web (no install) | web.chatboxai.app |
| iOS / Android | App Store / Google Play links from chatboxai.app |
| GitHub source / releases | github.com/chatboxai/chatbox/releases |
After installing, either enter your own API key for a cloud provider (OpenAI, Anthropic Claude, Google Gemini, Azure OpenAI) or connect a local server: install Ollama, pull a model (for example `ollama pull llama3.2`), and enter its local API base URL in Chatbox settings. You do not need a paid Chatbox AI plan to use local models -- the free app works fully offline against a local server.
Pricing: Free Core App or Paid Chatbox AI Plans
The Chatbox app is free with your own API keys or a local model server. Chatbox AI, the company's separate hosted service, is a paid product for people who would rather pay a flat monthly fee than manage cloud-provider API keys and billing themselves.
- Free (bring your own keys or local server): The full Chatbox app -- desktop, mobile, and web -- at no cost when you supply your own API key for OpenAI, Anthropic Claude, Google Gemini, or Azure OpenAI, or connect a local Ollama or LM Studio server. No account or subscription required for this path.
- Lite -- $3.99/month: 2,000,000 compute points, plus web search and image chat included in the plan, as of September 2026.
- Pro -- $19.99/month: 12,000,000 compute points, plus document chat and MCP integration on top of the Lite feature set.
- Pro+ -- $39.99/month: 30,000,000 compute points, the largest quota of the three paid tiers.
- Compute points consume at different rates depending on which model you select within a paid plan -- a larger or more expensive underlying model uses more points per message than a smaller one.
- Affiliate program: Chatbox is the only one of five compared chat-client apps (LobeChat, Big-AGI, NextChat, Page Assist, Chatbox) with a confirmed, live affiliate program, as of September 2026.
Chatbox vs LobeChat vs Big-AGI vs NextChat vs Page Assist: Which One?
All five are chat interfaces for AI models, but they split into two groups by deployment: Chatbox and Page Assist install directly as an end-user app or extension, while LobeChat, Big-AGI, and NextChat are self-hosted web apps you deploy to a server or Vercel.
Tool | Best for | Deployment | Price | License |
|---|---|---|---|---|
| Chatbox | One app across desktop, mobile, and web with an optional paid plan | Native install (Win/Mac/Linux/iOS/Android) + web | Free (own keys) / paid plans from $3.99mo | GPLv3 (core) |
| Page Assist | Chatting with local models and the current webpage, no server | Browser extension (Chrome, Firefox, Edge) | Free, donation-funded | MIT |
| LobeChat | Polished UI, agent builder, plugin marketplace | Self-hosted (Docker/Vercel/Zeabur) | Free self-hosted / Cloud from $9.9mo | LobeHub Community (Apache-based) |
| Big-AGI | Comparing several models on one prompt (Beam) | Self-hosted or hosted | Free self-hosted / Pro $10.99mo | MIT |
| NextChat | Fastest possible deploy, smallest footprint | Self-hosted (Docker/Vercel) or native app | Free self-hosted | MIT |
For a wider five-way look at these same apps, see LobeChat vs Big-AGI vs NextChat vs Page Assist vs Chatbox. For the full catalog, see the Local LLM Software Directory.
Who Should Use Chatbox?
Chatbox fits people who want one chat client that follows them across every device, with the choice of bringing their own API keys or paying a flat monthly fee instead.
- β You want the same app on your phone, laptop, and desktop -- native installs on five platforms plus a no-install web version all share one interface.
- β You would rather pay a flat monthly fee than manage API keys and billing across multiple providers -- a paid Chatbox AI plan bundles access behind one subscription.
- β You want to bring your own API keys and pay nothing to Chatbox -- the free app supports this fully, including local Ollama or LM Studio servers.
- β You want image generation and formatted output (Markdown, LaTeX, code) in the same window as your chat.
- β You want a plugin marketplace or agent builder as the primary feature -- see LobeChat instead.
- If unsure: install the free app, connect your own API key or a local Ollama model, and decide whether a paid Chatbox AI plan is worth it once you know your actual usage.
When Should You NOT Use Chatbox?
Cross-platform reach is Chatbox's main strength, but a few gaps are worth knowing before committing.
- If you want a plugin marketplace or an agent builder as the primary feature: LobeChat's plugin ecosystem and agent/persona builder are more developed.
- If you want to compare multiple models on the same prompt side by side: Big-AGI's Beam feature is built for exactly this; Chatbox has no built-in multi-model compare.
- If you specifically want a plain MIT license with no copyleft terms: Chatbox's core is GPLv3, which requires releasing modified redistributions under the same license -- see Page Assist, Big-AGI, or NextChat for MIT-licensed alternatives.
- If you want the lightest possible browser-only footprint with zero install: Page Assist runs as a browser extension rather than a separate native app.
- If you want a multi-user, team-managed deployment on your own infrastructure: Chatbox is a single-user client per account, not a self-hosted, multi-user server -- see Open WebUI for that model.
Competitors and Alternatives
Chatbox shares the "Web UIs & Browser Frontends" category in the Local LLM Software Directory with LobeChat, Big-AGI, NextChat, and Page Assist, but the five split into two genuinely different groups. Chatbox and Page Assist are both installable end-user clients -- an app or a browser extension you get directly, with no server to run -- while LobeChat, Big-AGI, and NextChat are self-hosted web apps you deploy yourself. The closest comparison for Chatbox is Page Assist, not the three self-hosted apps.
- Page Assist -- the other genuinely installable client in this group: a free, MIT-licensed browser extension rather than a native app, with no paid tier and no cross-platform mobile app.
- LobeChat -- a self-hosted, multi-provider chat interface with a plugin marketplace and agent builder, deployed to Docker or Vercel rather than installed directly.
- Big-AGI -- a self-hosted or hosted, MIT-licensed interface whose standout feature is comparing multiple models on one prompt at once (Beam).
- NextChat -- the lightest-weight self-hosted chat interface in the same directory segment, deployable with one click to Vercel.
Common Mistakes With Chatbox
- Assuming a paid Chatbox AI plan is required to use local models. It is not -- the free app connects to a local Ollama or LM Studio server at no cost, with no account needed.
- Confusing the free GPLv3 client with the paid Chatbox AI service. They share the same app, but Chatbox AI is a separate hosted product sold through Lite, Pro, and Pro+ subscriptions, not a requirement to use the software.
- Assuming GPLv3 works like the plain MIT license used by some competitors. Redistributing a modified version of Chatbox requires releasing your changes under GPLv3 as well -- check the license before building a commercial fork.
- Not entering a connection URL for a local server. Chatbox needs the local server's API base URL (for example, Ollama's default local address) entered in settings before it can reach it -- it does not auto-detect a local server on its own.
- Not checking the current release before relying on a specific feature. Chatbox is an actively developed project -- confirm the current feature set and pricing at github.com/chatboxai/chatbox and chatboxai.app before relying on it for a workflow.
Frequently Asked Questions
What is Chatbox?
Chatbox is an AI chat client with a GPLv3-licensed open-source core, distributed as native apps for Windows, macOS, Linux, iOS, and Android, plus a web version at web.chatboxai.app. The free app connects to your own API keys or a local server such as Ollama or LM Studio; the company also sells a separate hosted service, Chatbox AI.
Is Chatbox free?
The Chatbox app itself is free to download and use with your own API keys or a local model server. Chatbox AI, the hosted service that provides API access without your own keys, is a separate paid product with Lite ($3.99/month), Pro ($19.99/month), and Pro+ ($39.99/month) plans as of September 2026.
Is Chatbox open source?
The Chatbox client is licensed under GPLv3, with source code public at github.com/chatboxai/chatbox and 41,600+ GitHub stars. GPLv3 is a copyleft license: redistributing a modified version requires releasing your changes under GPLv3 as well, unlike a plain MIT license.
Does Chatbox work with local models like Ollama?
Yes. Chatbox connects to a local Ollama or LM Studio server by entering its API base URL in settings, alongside cloud providers such as OpenAI, Anthropic Claude, and Google Gemini, all configured with your own API keys.
What is the difference between the free Chatbox app and Chatbox AI?
The free Chatbox app is the GPLv3 client -- you supply your own API keys at no cost to Chatbox. Chatbox AI is the company's separate hosted service, sold through paid Lite, Pro, and Pro+ plans, that provides API access, web search, image chat, and document chat without requiring your own keys.
Does Chatbox store my conversations locally?
Yes. The Chatbox GitHub repository states that data remains on the device by default in the free app. Conversations synced through a Chatbox AI account are also stored on the company's servers as part of that separate hosted service.
Which platforms does Chatbox support?
Native apps for Windows, macOS, and Linux, plus mobile apps for iOS and Android, and a web version at web.chatboxai.app that needs no install -- six surfaces from one codebase.
Who maintains Chatbox?
Chatbox is built and maintained by its own team behind the Chatbox AI product, not an individual maintainer. It is not affiliated with LobeChat, Big-AGI, NextChat, or Page Assist -- five separate, unrelated projects.
How is Chatbox different from Page Assist?
Both are genuinely installable end-user clients rather than self-hosted server deploys. Page Assist is a free, MIT-licensed browser extension only, with no paid tier. Chatbox is a native desktop/mobile/web app under GPLv3, with an optional paid Chatbox AI hosted service on top of the free client.
Sources
- Chatbox source code, license, and README -- github.com/chatboxai/chatbox
- Chatbox AI official site, downloads, and pricing -- chatboxai.app
- Chatbox web app -- web.chatboxai.app
- LobeChat, Big-AGI, NextChat, Page Assist, Chatbox Compared -- PromptQuorum