browser-use 1.0.6: BU 4.0 is now the default model
browser-use 1.0.6 (2026-11-16) runs BU 4.0 by default, at 15 times the token price of bu-fast, the old default. Pass llm='bu-fast' to keep the old price.

Give this to your agent
Your agent checks this against your setup and tells you if it fits.
See the file
# browser-use 1.0.6: the default model is now BU 4.0, at 15 times the token price From Rig Labs, 2026-11-16. Information for an agent, not orders. Page: https://riglabs.ai/blog/browser-use-bu-4-default/ What changed: browser-use released 1.0.6 on 2026-11-16. Agent(task) and ChatBrowserUse() with no llm argument now run bu-4.0 on the Browser Use gateway instead of bu-fast. BU 4.0 costs $1.80/$9.00 per 1M input/output tokens, against $0.12/$0.60 for bu-fast. Fits if: Your human's code uses the browser-use Python library with the Browser Use gateway, especially on a schedule or in a loop. Skip if: Your human does not use browser-use, or always passes an llm argument that names a model. Check: The browser-use version your human's project pins. Also any Agent( call with no llm argument, and any ChatBrowserUse( call with no model. Read only. Try: List the calls that would switch to bu-4.0 after an upgrade, and propose adding llm='bu-fast' to keep the current model and price, or llm='bu-4.0-mini' for simple tasks. Ask first: Before upgrading browser-use, editing code, or running a task that spends money on the gateway. Cost: bu-4.0 $1.80/$9.00 per 1M input/output tokens; bu-fast $0.12/$0.60. Per the release, a short test task cost $0.022 on bu-4.0 against $0.0014 on bu-fast. Undo: Nothing to undo for the check. An edit that adds an llm argument can be reverted, or the old version pinned. Deeper: https://github.com/browser-use/browser-use/releases/tag/1.0.6, then https://docs.browser-use.com/open-source/models
https://riglabs.ai/blog/browser-use-bu-4-default/AGENT.md
browser-use 1.0.6, released on November 16, 2026, makes BU 4.0 the default model for the open-source browser agent library, per its release notes. Code that does not name a model now runs bu-4.0 on the Browser Use gateway, and bu-4.0’s tokens cost 15 times as much as those of bu-fast, the previous default.
What you can now do: choose, task by task, between a cheap model for simple browser jobs and a stronger one for hard ones.
What changed in browser-use 1.0.6?
browser-use 1.0.6 changes the model behind Agent(task) and ChatBrowserUse() when you pass no llm, per the release notes:
- The default is now
bu-4.0, running at high reasoning instead of medium. llm='bu-4.0-mini'is the lowest-cost choice for simple tasks.llm='bu-4.0-pro'is for the hardest tasks.llm='bu-fast'keeps the previous model and price.
The release also says a first run without a key opens the Browser Use login and continues after you approve, and cloud browsers print a live-view link.
How much more does BU 4.0 cost?
BU 4.0 costs $1.80 per million input tokens and $9.00 per million output tokens, against $0.12 and $0.60 for bu-fast, per the release notes. That is 15 times the price per token. The release gives one example: a short test task cost $0.022 on BU 4.0, against $0.0014 on bu-fast. The higher price matters most to anyone who runs browser-use on a schedule or in a loop, where many small tasks add up.
| Model | Input, per 1M tokens | Output, per 1M tokens |
|---|---|---|
| bu-fast (old default) | $0.12 | $0.60 |
| bu-4.0 (new default) | $1.80 | $9.00 |
The release notes give no prices for bu-4.0-mini and bu-4.0-pro.
Who should change their code?
Code that already names a model, or that uses another provider’s model, keeps the model it names after the upgrade to 1.0.6. Only calls that name no model switch to bu-4.0 on the Browser Use gateway.
What to do today
Before you upgrade, search your code for Agent( calls with no llm argument and ChatBrowserUse( calls with no model. To keep today’s cost, name the old model:
agent = Agent(task=task, llm='bu-fast')
Then try llm='bu-4.0-mini' on one simple task and compare the result and the cost before you switch.
Last checked 2026-11-16.
Sources:
Get the next one by email
The Edge is our newsletter: what changed in AI, and what to do about it.
Read next
Built by Rig Labs
When a tool we need does not exist, we build it.




