Settings

Theme

PA bench: Evaluating web agents on real world personal assistant workflows

vibrantlabs.com

37 points by shahules a day ago · 10 comments · 1 min read

Reader

We’re the team at Vibrant Labs (W24). We’ve been building envs for browser agents and quickly realized that existing benchmarks in this space didn’t capture the primary failure modes we were seeing in production (which scaled up as the number of applications and horizon length increase).

We built PA Bench (Personal Assistant Benchmark) to evaluate frontier computer/web use models on their ability to handle multi-step workflows across simulated clones of Gmail and Calendar.

*What’s next:*

We’re currently scaling the dataset to 3+ tabs and are building more high-fidelity simulations for common enterprise workflows. We’d love to hear feedback on the benchmark and notes about what was/wasn’t surprising about the results.

Blog post: https://vibrantlabs.com/blog/pa-bench

mrorigo 11 hours ago

I just don't get why would you would want an agent to use the browser to do these mundane things (check email, work with calendar etc), when you can simply give it a few tools, and save maybe six gazillion tokens per task?

  • shenberg 7 hours ago

    Using existing enterprise apps probably - this solution is scalable for the vendor and it's easier to sell using existing software as-is than to start out by writing new custom tools.

    • shahulesOP 2 hours ago

      After doing few experiments, I think that having Agents work on browser for all tasks wouldn't be best due to many factors like token cost, safety, etc. But browser/computer can be a tool that the agent can be alongside MCPs to complete tasks that requires interaction with such modalities.

abhijithneil 21 hours ago

Is there a possible way computer use can be automated using multiple computer use agents from different providers, but also with some sort of routing setup so the best course of action can be chosen without hitting failures (for eg: permission issues in OpenAI could be rerouted to Gemini)

  • shahulesOP 20 hours ago

    There are few agents like browser-use, skyvern etc that may provide this capability.

AIorNot 10 hours ago

Well if these guys computer action model works as they intended (ground up video trained model)

https://news.ycombinator.com/item?id=47125014

maybe this benchmark will be conquered far faster then expected

Keyboard Shortcuts

j
Next item
k
Previous item
o / Enter
Open selected item
?
Show this help
Esc
Close modal / clear selection