@Stagehanddev

the SDK for browser agents, built by @browserbase join the discord: https://xitter.cf/t.co/Q5UxzlgTft

 chrome devtools
stagehand.dev
Joined February 2025
  • Tweets570
  • Following47
  • Followers5.6K
88Media
Introducing our Browser Agent Evals. We benchmark frontier and open models using various harnesses on computer-use tasks. Our evals are fully open source and reproducible via our Evals CLI.
10
11
3
39
7,315
You can extract typed data from any website using @pydantic and Stagehand. Using Pydantic's enforced typing and Stagehand's structured extract, you can build robust and self-healing web scraping pipelines. Read our full blog post below.
3
3
9
501
Read the full post: stagehand.dev/blog/pydantic-…

Extract typed data from any website using Pydantic and Stagehand

Learn how to combine Pydantic and Stagehand to reliably extract typed data from any website using only 50 lines of code.

stagehand.dev
1
53
Power your browser agents with Stagehand v4 in OpenAI's official SDKs: docs.stagehand.dev/v4/integr…

OpenAI Toolset - Stagehand

Give GPT browser tools through the OpenAI SDK, with Stagehand driving local Chrome or a Browserbase session.

docs.stagehand.dev
Kevin Whinnery
@kevinwhinneryOct 16
Browser and computer use got waaaay easier in the latest @OpenAIDevs Python and Node.js SDKs 🧵
5
5
18
815
Full repo & examples here: github.com/browserbase/opena…

GitHub - browserbase/openai-cua-toolset: Give OpenAI models a browser through the OpenAI SDK's...

Give OpenAI models a browser through the OpenAI SDK's browser toolset, with Stagehand as the driver - browserbase/openai-cua-toolset

github.com
2
119
Stagehand retweeted
Shrey Pandya
@shreypandyaOct 16
New ship on browse.sh - record mode! -> fire up a browser session -> show your agent what to do, click-by-click -> hit export to download your finished workflow you get a SKILL.md your agent can follow, plus scripts for @Stagehanddev, Playwright, or Puppeteer.
4
3
28
1,343
Stagehand retweeted
Browserbase
@browserbaseOct 14
Browserbase Search is now powered by Parallel (@p0) as the default search provider. Parallel Fast returns sub-second results when speed matters. Parallel Advanced goes deeper for multi-step research when quality matters most.
20
19
10
105
29,032
Gemini 4 Argon is live in Stagehand. It outperforms Fable 5.1 on computer use tasks with @eve and @pidotdev harnesses, while being roughly around the same cost/speed. You can try the model in our playground at gemini.browserbase.com.
2
3
9
699
See our benchmarks: stagehand.dev/evals

Browser Agent Evals

How today's models perform on computer use benchmarks, compared on accuracy, cost per task, and wall-clock speed.

stagehand.dev
3
116
Stagehand retweeted
Browserbase
@browserbaseOct 12
We're sponsoring the AI Assistant meetup on Monday October 19th @ 7:30pm. Come meet other builders and hear from teams building the infra powering the fastest growing assistants on the market. Link to RSVP below.
David Pawlan
@DavidPawlanOct 12
Pumped to announce 4 more sponsors for the SF AI Assistant meetup! > @thelinqapp > @browserbase > @crossmint > @mem0ai RSVP: luma.com/wabd65c0 More details on the agenda to come
1
1
12
1,011
Stagehand retweeted
Shrey Pandya
@shreypandyaOct 12
Coolest new feature in Stagehand - we can now automatically detect & invoke WebMCP tools on the fly. This unlocks interacting with all @stripe checkout pages without repeatedly polling for new tools. Give it a try on 4.2!
Stagehand
@StagehanddevOct 12

Introducing Stagehand 4.2: WebMCP hooks, PDF exports, and more

Article · @Stagehanddev
6
8
1
38
2,811

Introducing Stagehand 4.2: WebMCP hooks, PDF exports, and more

Article · @Stagehanddev
5
1
23
3,171
Claude 5.5 Haiku is on par with GPT 6 Luna from OpenAI on accuracy and speed, but is 3x the cost per task. As computer use improves, most gains will show up on the Pareto frontier: maintaining high accuracy while also being faster and cheaper.
2
2
10
542
Find out how well your favorite models can use a computer: stagehand.dev/evals

Browser Agent Evals

How today's models perform on computer use benchmarks, compared on accuracy, cost per task, and wall-clock speed.

stagehand.dev
1
133
Stagehand retweeted
Charly Poly
@whereischarlyOct 8

6 ways to translate a web page into LLM context using Python

Article · @whereischarly
1
2
11
1,266
Stagehand supports @ClaudeDevs' new Tool SDK, converting model actions into browser actions. Follow our quickstart guide to build your first browser agent with Claude and Stagehand. github.com/browserbase/claud…
ClaudeDevs
@ClaudeDevsOct 7
Computer use and browser use toolsets are now built into the Python and TypeScript SDKs for Claude. The API tells you what Claude wants to click or type. Previously, you had to write your own loop and map clicks and keystrokes to commands, but the SDKs now run the loop and send actions to drivers.
3
3
2
21
4,775
Here are our docs to get started: docs.stagehand.dev/v4/integr…

Claude Toolset - Stagehand

Give Claude browser tools through the Anthropic SDK, with Stagehand driving local Chrome or a Browserbase session.

docs.stagehand.dev
147
Computer use models are getting really fast. Opus 5 (released in July '26) averages 678s per task while newer models like Opus 5.5 average 326s, ~2x speed up. See the speed up for yourself across 72 model x harness combinations at stagehand.dev/evals
4
4
20
1,381
Stagehand retweeted
Browserbase
@browserbaseSep 25
How do you actually build an effective harness with Claude? We had Thariq (@trq212) from Anthropic at Navigate 2026 to talk about "Unhobbling Claude", the difficulties and processes on how to build agents and harnesses.
7
5
1
51
11,686
GPT-6 Luna is a step function improvement from it's predecessor scoring 11% higher in accuracy on our browser agent tasks. GPT-6 Sol scores 2% better in accuracy, but costs 20x more per task than Luna.
OpenAI Developers
@OpenAIDevsSep 22
GPT-6 Sol and Luna just landed in Astra’s orbit. Both launch today with API prices 50% lower than GPT-5.6. Build with Sol. Scale with Luna. To production and beyond.
6
2
1
17
1,626
Checkout our evals: stagehand.dev/evals

Browser Agent Evals

How today's models perform on computer use benchmarks, compared on accuracy, cost per task, and wall-clock speed.

stagehand.dev
2
290
Load more