BrowserOS 이전 버전 및 전체 버전

✨ What's new
- No BrowserOS account needed. Sign-in and cloud-sync screens have been removed. Conversations continue to stay on your device, and sidebar history is now available to everyone.
- Connect your preferred AI. The built-in BrowserOS model has been retired. Connect an LLM provider or a coding agent in AI Settings to chat. If a scheduled task used the built-in model, choose a connected provider for it too.
- An updated model picker. The ChatGPT subscription model list now includes newer model families, and the provider catalogue has been refreshed. You can also enter a model ID directly; availability depends on your provider and subscription.
- Screenshots reach custom providers. Fixed an issue that prevented vision-capable models on OpenRouter and other OpenAI-compatible providers from receiving the agent's screenshots.
- Chromium 155. Updated to Chromium 155.0.8059.26.

✨ What's new
- A simpler BrowserOS agent. We've redesigned the agent experience to make it easier to use, with a cleaner interface and more straightforward setup. Your conversation now follows you across the tabs the agent works in.
- Better support for your AI provider. Improved compatibility across supported providers, including OpenAI, Anthropic, Google, and OpenRouter. Model selection and reasoning settings now better reflect what each model supports, with clearer error messages when something goes wrong.
- A much lighter agent extension. We cut the unpacked extension size by 86%, from 42.93 MB to 6.02 MB, so there's much less for the browser to unpack at startup.
- Chromium 151. Updated to Chromium 151.0.7922.137, bringing upstream security fixes and browser improvements.
Also changed: Voice input has been removed in this release.

✨ What's new:
- Auto-updates on Windows 🎉 — Long pending, finally here! BrowserOS now updates itself on Windows, just like on macOS. No more downloading and manually installing every release
- Fixed the recurring Windows pop-up — Squashed the bug where a dialog kept popping up every ~15 minutes on Windows, and improved its reliability overall. Please let us know if you still see it!
- Third-party agent integrations — Fixed a bunch of bugs and issues with agent integrations like Codex and Claude Code — connecting and staying connected is now much more reliable.

✨ What's new:
- Run Claude Code & Codex right in your browser — We've extended the agent harness to bring full coding agents into BrowserOS. Claude Code and Codex now come bundled and plug straight into the assistant, so you can drive your browser with the agent — and the subscription — you already use.
- A brand new experience — A redesigned new tab, a calmer composer, and a rebuilt command center for switching between agents. The whole assistant is cleaner, faster to reach, and easier to live in.
- New MCP tools — We rebuilt the browser tool surface from the ground up — a tighter, more reliable set of tools for agents to drive the browser. Plus one-click install of BrowserOS as an MCP server into the agents you already run, with automatic URL sync.
- Chromium 148 — Updated to the latest Chromium base with all recent upstream fixes and security patches.
- Streamlined — We've pulled back a few features that weren't getting much use — Skills, Soul, and Memory — so we can focus and ship better versions of them soon.

✨ What's new:
ChatGPT Pro & GitHub Copilot — Already paying for ChatGPT Pro, Plus, or GitHub Copilot? Now you can use that subscription to automate browser tasks in BrowserOS — at no extra cost. Just log in with your existing account.
Voice Input — Talk to your agent instead of typing. The chat bar now accepts voice input for any query.
Smarter Agent — Overhauled agent grounding for more reliable and precise browser actions.
Bug Fixes & Reliability — Improved memory tools, fixed fill tool issues, new tab layout fixes, and numerous stability improvements across the board.

✨ What's new:
Skills — Define reusable instructions that teach your agent how to handle specific tasks. Ships with 12 built-in skills like deep research, price comparison, and form filling — and you can create your own. Docs →
Memory — Your assistant now remembers you across conversations. Core memory stores permanent facts (your name, projects, preferences), while daily notes capture session context and auto-expire after 30 days. Everything stays local on your machine. Docs →
Kimi K2.5 — in partnership with Moonshot AI — Kimi K2.5 is now the default model in BrowserOS. Open-source, multimodal, and great for browser automation. For a limited time, users get extended usage limits at no cost. Bring your own key from platform.moonshot.ai anytime.
SOUL.md — Shape your assistant's personality, tone, and boundaries. It evolves on its own as you interact — the more you use it, the more it feels like yours. Docs →

✨ What's new:
- SOUL.md — Your assistant now has a soul. Tell it how you like to communicate, set boundaries, shape its personality — and it adapts on its own over time. The more you use it, the more it feels like your assistant. Inspired by OpenClaw — thanks to @steipete for the great concept. Docs →
- Vertical tabs — One of the most requested features is here. BrowserOS now ships with vertical tabs by default. More screen space, better tab management, and a cleaner layout out of the box. Prefer horizontal? You can switch back anytime in settings. Docs →
- Long-term memory — Your assistant finally remembers you. Your name, your projects, what you talked about last week — it carries context across every conversation so you never have to repeat yourself. All stored locally on your machine. Also inspired by OpenClaw. Docs →
- Chromium 146 — Updated to the latest Chromium release with all recent upstream fixes and security patches.

✨ What's new:
- New agent (v3) — Completely redone from scratch. 2x faster, 2–3x better performance.
- Tools — major upgrade — Agent tools and MCP server both got a big overhaul. ~20 new tools (54 total) including file upload, save as PDF, background windows, and more. Connection with third-party coding agents (Claude Code, Codex, etc.) is much better now.
- General fixes — Better agent installation, bug fixes, and smoother experience overall.
- Linux Debian packaging — Fixed the remaining Debian packaging issues.

✨ What's new:
- Chromium 145 — Upgraded to the latest Chromium base with all recent upstream fixes and security patches.
- Login session import improvements — Importing login sessions is now more reliable.
- Stability & reliability — General improvements across the board.

✨ What's new:
- Sync — Save your browser configuration, agent history, and scheduled tasks across machines. Your setup now follows you wherever you go.
- App Connector redesign — Connecting MCP apps is now easier. The new App Connector UI makes it simpler to discover, install, and manage your connected apps.
- MCP port stability — Additional fixes for users still experiencing port issues. More reliable connections across restarts.

✨ What's new:
- MCP port fix on Windows & Linux — Fixed an issue with MCP port. Port now stays consistent across restarts and made it more reliable.
- Settings fix — chrome://browseros/settings now works correctly.
- Improved agent — General reliability and performance improvements for the agent.

✨ What's new:
- Workflows — Build reliable, repeatable browser automations with a visual graph builder. Chat with the workflow agent to define step-by-step automation—ideal for complex tasks where ad-hoc prompts aren't enough. Read more →
- Workspaces — Give the agent access to a folder on your computer. Combine browser automation with local file operations—research on the web and save reports, scrape data and export to files, all in one task. The agent is sandboxed to your selected folder only. Read more →

🔧 Bug fix release
MCP server disconnect fix — Some users were experiencing MCP server disconnects during use. This release fixes the underlying port handling issue. If you've been seeing your MCP connections drop unexpectedly, please upgrade!

📝 Note: This is a minor release focused on stability. If you're not experiencing MCP issues, you can skip this update.

BrowserOS - v0.36.0
✨ What's new:
- Agent personalization — Add your own prompts to personalize the agent. Tweak its behavior, adjust how it responds, set your preferred formatting, and more.
- Toolbar customization — Hide the Hub chat and labels from the settings page to declutter your toolbar.
- MCP server port stability — The port now stays consistent through browser restarts, so you don't have to keep updating your MCP clients with a new port.
- Fixed agent install/update issues — We've made the startup experience much smoother. The agent now handles installs and updates more proactively instead of only on the first launch.

📝 Note for Windows & Linux users: Please download v0.36.1 — this is the latest release and what the update notification is referring to. Sorry for the inconvenience!

BrowserOS - v0.35.0
Merry Christmas from the BrowserOS team! 🎄🎁
✨ What's new:
- Agent stability fixes — We fixed a few bugs to make the agent loop much more reliable.
- Gemini 3 support — Gemini 3 is now supported both through openrouter and google LLM adapter.
- Better error surfacing — Error messages are now clearer, so you can actually tell what went wrong.
Happy holidays! 🎅

BrowserOS v0.34.0
✨ What's new:
- Third-party MCP server support — Connect external MCP servers like Google Calendar, Notion, Google Docs, Gmail, and more to your BrowserOS agent. You can also connect your own custom MCP servers!
- Gemini 3 support — Both Gemini 3 Pro and Flash models now work with BrowserOS.
- Windows icon fix — The Windows icon now displays BrowserOS logo correctly.
- Agent & UI improvements — Various agent loop fixes and UI polish to make everything smoother.
🎬 Check out a short video on how to connect MCP servers to BrowserOS!
Give it a try and let us know what you think 🙌

BrowserOS - v0.33.0
✨ What's new:
- OpenAI-compatible provider support — You can now connect any OpenAI-compatible API endpoint in providers.
- Multi-window & multi-profile agent support — Agent now works smoothly across multiple windows and browser profiles, with affinity to the profile/window where it was started.
- MCP server reliability — Fixed connection drops and improved stability.
- Agent reliability improvements — General stability fixes to make the agent more reliable.

A Brand New BrowserOS!
We've been heads down building, and we're SO excited to share this with you — this is a complete revamp of BrowserOS! 🎉
✨ What's new:
- New Agent — Meet the completely rebuilt agent! Faster, smarter, and way more reliable.
- Agent Per Tab — You can now spin up an agent in each tab. Run multiple agents doing different things, all at the same time.
- Polished UI — We've given the entire interface a fresh coat of paint. It's cleaner, more refined, and just feels better.
- Manifest V2 Support — this means you can install ad blockers like uBlock Origin again!
- Native Split View — Work on two things side by side with a simple shortcut.
- Chromium 142 — We're now running on the latest Chromium (142.0.7444.49).
- Azure & AWS Bedrock — Added native support for Azure OpenAI and AWS Bedrock models.
⚠️ Breaking Changes:
- Old agent will stop working — please update to the new browserOS version asap.
- LLM provider settings won't carry over — You'll need to reconfigure your API keys and model settings. Sorry about that!
https://github.com/user-attachments/assets/39c9fdb2-5035-47f8-a933-92a4151a66c2

🐶 What's new:
- Added better text extraction feature in LLM Chat (when you press the copy button)
- Several bug fixes:
- Improved MCP server support wrt to stability of connection
- Unpinned few of the extensions for cleaner look
- Fixed support for third-party MCP servers with new agent

✨ What's new:
- Our browsing agent just got a major upgrade. New agent is now much more reliable at handling complex tasks, powered by our new MCP tools.
- Improved MCP tools support for better reliability. no more random disconnects requiring browser restarts.
- Will be rolling out the new Agent to BYOK mode in upcoming releases.

✨ What's new:
- Adding support for configuring providers in LLM Hub and LLM Chat. Navigate to chrome://settings/browseros to configure it!
- Other bug fixes and improvement in Agent.

✨ What's new:
- Debian Linux support – Now available as a .deb package for Debian-based systems.
- Custom LLM endpoints – Configure custom URLs for LLMChat and LLMHub to use your preferred endpoints. Will be available as option soon in our new settings page.
- Agent stability fix – Resolved the issue where agents were getting disabled for some users. (#93)
- Bug fixes and improvements.

✨ What's new:
- Bug fixes and improvements.
- Created detailed docs at https://docs.browseros.com/
- Introduced a more seamless LLM settings page, thanks to @OmkarBansod02 for the great work! The new LLM provider settings page allows you to quickly test the BrowserOS->LLM provider connection and provides a more seamless way to add different providers -- Ollama, LMStudio or API keys. (screenshot below)

BrowserOS v0.25.0 - Chromium 137.0.7209.69

✨ What's new:
- Agent v3 is here! – Our browsing agent just got a major upgrade to v3, now much smarter at handling complex tasks and workflows.
- Enhanced local model support – Now using combined Visual + semantic DOM for better performance with local models.
- UI improvements – Various interface enhancements. Thanks to @OmkarBansod02 for the contributions!

✨ What's new:
- Improved click reliability in our C++ APIs.
- Other fixes and improvements.

What's new:
- We have launched MCP support for popular services! Thanks to Zihao from Klavis.ai for helping us with integration!
- We have improved the extraction tool, thanks to the DFS parser implemented by Omkar Bansod.
Try the MCP integration and let us know your feedback! You should get the agent update OTA soon or you can force update the agent!

What's new
- New updated Agent UI!
- Fixed MacOS bug (#48) which caused the app to crash on startup for some users. This unfortunately also makes a breaking change, requiring re-installation of extensions and logins.

Release v0.19.0

- New revamped Agent infra! This should improve our agents reliability and performance. Please test it out and give us your feedback!

- Updates to our Agent, rolling out new Agent infra.

- Fixes #37 bug where we shared still some paths with chromium.
- Few other performance improvements and bug fixes.

- Support for screenshot in LLM chat.
- New settings page chrome://settings/browseros-settings for configuring BrowserOS internals. Today, it supports hiding labels in toolbar.
- Various fixes and improvements.

What's New in v0.14.0
- Reverting to older agent; people reported few issues with new one. Will get that fixed and roll it out!
- Performance improvements and UX fixes.
Downloads
- macOS: Universal, ARM64, and x64 builds available
- Windows: x64 installer (exe and zip)
- Linux: x64 AppImage

- New Agent infra - Increases the reliability of our agent
- Fixes to Agent UI - Light mode support

- New Clash of GPTs!
Run up to three LLMs side-by-side in a single window, so you can send one prompt and instantly compare their answers without ever switching windows.