diff --git a/README.md b/README.md
index 587ff15aa..8a9029395 100644
--- a/README.md
+++ b/README.md
@@ -47,7 +47,7 @@ Browser Use gives AI agents a browser. Describe a task, and the agent navigates
Find an available slot, handle the CAPTCHA, and schedule the test with Browser Use V4.
-[](https://x.com/mathisdittrich/status/2078619618265141560)
+
[Watch Johannes's demo ↗](https://x.com/mathisdittrich/status/2078619618265141560)
@@ -57,61 +57,18 @@ Find an available slot, handle the CAPTCHA, and schedule the test with Browser U
# Which Browser Use do I need?
-
-
-
-
-
-
-
Your agent, your browser
-
Connect your coding agent with the CLI, or build automation with the Python library.
-
-
+
-
-View the diagram vertically
+- **[CLI](#cli-quickstart):** Give Claude Code, Codex, OpenCode, Pi, or another agent browser access.
+- **[Python library](#python-library):** Run the Browser Use agent locally from your own code.
+- **[Fully hosted cloud](#fully-hosted-cloud):** Send a task through the API; we run the agent and browser.
-
-
-
-
+## CLI quickstart
-
-
-[How the pieces fit together ↗](https://docs.browser-use.com/cloud/which-product)
-
-Eligible new Google, GitHub or Microsoft signups get a one-time **$15 Cloud credit**. No card required; email/password signups do not qualify. [Pricing and eligibility](https://browser-use.com/pricing.md).
-
-# Quickstart
-
-If you want to use Browser Use in your agent (Claude Code, Codex, OpenCode, Cursor, Hermes, OpenClaw, etc.), paste this prompt, and it sets everything up itself:
+If you want to use Browser Use in your agent (Claude Code, Codex, OpenCode, Pi, Cursor, Hermes, OpenClaw, etc.), paste this prompt, and it sets everything up itself:
```text
Install or upgrade browser-use to the latest stable version with uv using Python 3.12, run `browser-use skill install` to register the skill, and connect it to my browser. If setup or connection fails, follow https://github.com/browser-use/browser-harness/blob/main/install.md.
@@ -121,9 +78,9 @@ Then tell your agent what you want done.
-# Python library: the easiest way to automate the web
+## Python library
-Want to automate the web at scale, from your own code, and with any LLM? Use the Python library:
+Run the Browser Use agent locally from Python, with your choice of model and a local or cloud browser:
**1. Install Browser Use (Python >= 3.11):**
@@ -160,7 +117,17 @@ if __name__ == "__main__":
asyncio.run(main())
```
-Check out the [library docs](https://docs.browser-use.com/open-source/introduction) and the [cloud docs](https://docs.browser-use.com/cloud/quickstart) for more!
+[Python library docs ↗](https://docs.browser-use.com/open-source/introduction)
+
+
+
+## Fully hosted cloud
+
+Send a task to our hosted V4 agent. We run the agent, browser, and infrastructure.
+
+[Get started with the API ↗](https://docs.browser-use.com/cloud/agent/quickstart)
+
+New Google, GitHub, or Microsoft signups get **$15 cloud credit**. No card required.
@@ -172,30 +139,6 @@ This [very hard benchmark](https://github.com/browser-use/benchmark) targets the
Browser Use is also **#1 on the [Odysseys leaderboard](https://odysseysbench.com/leaderboard)** with an 87.4% average, ahead of computer-use agents from OpenAI, Anthropic, Google, and Microsoft. Odysseys measures the agent's performance on 200 long-horizon web tasks.
-# Open Source vs Cloud
-
-**Use the Open-Source Agent**
-- Free, and runs on your own machine
-- Deep code-level integration and control: pick your LLM, customize the agent's behavior
-- We recommend pairing it with our [cloud browsers](https://docs.browser-use.com/open-source/customize/browser/remote) for leading stealth, proxy rotation, and scaling
-
-**Use the [Fully-Hosted Cloud Agent](https://cloud.browser-use.com?utm_source=github&utm_medium=readme-hosted-agent) (recommended)**
-- Much more powerful agent for complex tasks
-- Easiest way to start and scale
-- Best stealth with proxy rotation and captcha solving
-- 1000+ integrations (Gmail, Slack, Notion, and more)
-- Persistent filesystem and memory
-- Rerunnable scripts fetch live data, even when sites change ([guide](https://docs.browser-use.com/cloud/agent/scripts))
-
-```sh
-curl -X POST https://api.browser-use.com/api/v4/runs \
- -H "X-Browser-Use-API-Key: $BROWSER_USE_API_KEY" \
- -H "Content-Type: application/json" \
- -d '{"task": "Your task"}'
-```
-
-
-
## Integrations, hosting, custom tools, MCP, and more on our [Docs ↗](https://docs.browser-use.com)
@@ -205,7 +148,7 @@ curl -X POST https://api.browser-use.com/api/v4/runs \
Should I use the CLI vs. the Python library?
-**Use the CLI** if you already have an agent (Claude Code, Codex, Cursor, Hermes, OpenClaw, etc.) that you want to complete browser tasks for you. The agent installs the skill once (see [Quickstart](#quickstart)) and can then control the browser. Examples:
+**Use the CLI** if you already have an agent (Claude Code, Codex, OpenCode, Pi, Cursor, Hermes, OpenClaw, etc.) that you want to complete browser tasks for you. The agent installs the skill once (see [CLI quickstart](#cli-quickstart)) and can then control the browser. Examples:
- "Upload this video to YouTube"
- "Compare these three laptops and give me a table with prices"
- "Fill in this job application with my resume"
diff --git a/static/readme/README.md b/static/readme/README.md
index a93d46850..0e0eddfe0 100644
--- a/static/readme/README.md
+++ b/static/readme/README.md
@@ -1,36 +1,27 @@
-# README artwork and demo sources
+# README diagram and demo sources
-These assets reuse Browser Use's existing website and documentation artwork.
-No demo footage was generated for this README.
+## Product diagram
-## Product artwork
+`which-product-light.svg` and `which-product-dark.svg` are editable SVGs.
+They are adapted from [the SDK product diagram](https://github.com/browser-use/sdk/blob/bfc0140801caf6a4c518c811f03674135efbd4aa/docs/cloud/images/which-product-light.svg),
+introduced in [SDK PR #241](https://github.com/browser-use/sdk/pull/241).
-Copied unchanged from the [Browser Use website](https://browser-use.com),
-`browser-use/website` commit `f07970a`:
+The README version shows three paths:
-| Local file | Website source |
-| --- | --- |
-| `local-agent.jpg` | `/lander/plates/oss-14b77bfd.jpg` |
-| `cloud-browsers.jpg` | `/lander/plates/browsers-8dd60aa0.jpg` |
-| `hosted-agent.jpg` | `/lander/plates/agents-be1dd941.jpg` |
+- Fully hosted cloud: OpenCode → Browser Use CLI → cloud browser.
+- CLI: your existing agent → Browser Use CLI → local or cloud browser.
+- Python library: your code → Browser Use agent → local or cloud browser.
-## Product map
+Keep the light and dark SVGs structurally identical when changing the diagram.
+The SDK source remains the reference for the hosted stack; this README adds the
+Python-library path and omits the Playwright branch.
-The four `which-product-*.svg` files are copied unchanged from
-[`browser-use/sdk`, `docs/cloud/images`](https://github.com/browser-use/sdk/tree/main/docs/cloud/images).
-They were introduced in [SDK PR #241](https://github.com/browser-use/sdk/pull/241).
-Keep the local/cloud routes and separate Profile Sync path aligned with the
-[product guide](https://docs.browser-use.com/cloud/which-product).
+## Driving-test demo
-## Demo thumbnail
+The inline GIF is converted from [Johannes Dittrich's public driving-test video](https://x.com/mathisdittrich/status/2078619618265141560).
+It preserves the full 20.4-second recording at 960×540, 10 frames per second,
+with an infinite loop. The source recording already obscures the contact fields.
+No demo footage was generated.
-| Local file | Published source |
-| --- | --- |
-| `driving-test.jpg` | [Johannes Dittrich's driving-test demo](https://x.com/mathisdittrich/status/2078619618265141560) |
-
-The README links to the playable public recording. Keep the poster and
-caption consistent with that recording when updating them.
-
-`driving-test.jpg` is the 8-second frame of Johannes's public recording,
-showing the appointment-location selection. The personal contact fields are
-not in this frame.
+The GIF is hosted as a [GitHub attachment](https://github.com/user-attachments/assets/135885e8-1141-4e10-b719-bf690ae7d260)
+so cloning the repository does not download the animation.
diff --git a/static/readme/cloud-browsers.jpg b/static/readme/cloud-browsers.jpg
deleted file mode 100644
index 116475b99..000000000
Binary files a/static/readme/cloud-browsers.jpg and /dev/null differ
diff --git a/static/readme/driving-test.jpg b/static/readme/driving-test.jpg
deleted file mode 100644
index 54a142e62..000000000
Binary files a/static/readme/driving-test.jpg and /dev/null differ
diff --git a/static/readme/hosted-agent.jpg b/static/readme/hosted-agent.jpg
deleted file mode 100644
index 94635ec35..000000000
Binary files a/static/readme/hosted-agent.jpg and /dev/null differ
diff --git a/static/readme/local-agent.jpg b/static/readme/local-agent.jpg
deleted file mode 100644
index c76661729..000000000
Binary files a/static/readme/local-agent.jpg and /dev/null differ
diff --git a/static/readme/which-product-dark-mobile.svg b/static/readme/which-product-dark-mobile.svg
deleted file mode 100644
index 37479eadb..000000000
--- a/static/readme/which-product-dark-mobile.svg
+++ /dev/null
@@ -1,56 +0,0 @@
-
diff --git a/static/readme/which-product-dark.svg b/static/readme/which-product-dark.svg
index e6bc8034c..f2ae5d745 100644
--- a/static/readme/which-product-dark.svg
+++ b/static/readme/which-product-dark.svg
@@ -1,51 +1,86 @@
-
+
diff --git a/static/readme/which-product-light-mobile.svg b/static/readme/which-product-light-mobile.svg
deleted file mode 100644
index 72811dff5..000000000
--- a/static/readme/which-product-light-mobile.svg
+++ /dev/null
@@ -1,56 +0,0 @@
-
diff --git a/static/readme/which-product-light.svg b/static/readme/which-product-light.svg
index 86429d589..4ae48a20c 100644
--- a/static/readme/which-product-light.svg
+++ b/static/readme/which-product-light.svg
@@ -1,51 +1,86 @@
-
+