Registry / ai-ml / webarena-verified

webarena-verified

JSON →
library1.2.3pypypiunverified

A verified version of the WebArena Benchmark for evaluating web agent capabilities. Current version 1.2.3, updated as of early 2025. Follows a quarterly release cadence.

pip install webarena-verified
INSTALL
IMPORT
SIG · WEBARENA-VERIFIED
W
webarena-verified
ai-mlpythonv1.2.3
harness data pending
Install & Compatibility
Where this runs

No compatibility data collected yet for this library.

Code
Verified usage

Verified import paths — ran on the pinned version, not inferred.

WebArenaEnv
from webarena_verified import WebArenaEnv
from webarena import WebArenaEnv
The package uses underscore in module name.
run_episode
from webarena_verified import run_episode
from webarena_verified.evaluate import run_episode
run_episode is top-level, not in a submodule.
BrowserManager
from webarena_verified.utils import BrowserManager
from webarena_verified.browser import BrowserManager
BrowserManager moved to utils in v1.2.0.

Initialize environment, run a sample episode, and clean up.

import os from webarena_verified import WebArenaEnv, run_episode config = { "browser": "chromium", "headless": True, "openai_api_key": os.environ.get('OPENAI_API_KEY', ''), } env = WebArenaEnv(config=config) result = run_episode(env, task="Find the cheapest flight from NYC to London") print(result) env.close()
Debug
Known issues
breakingv1.2.0 renamed the module from 'webarena' to 'webarena_verified'. Old imports will fail.
fix
Change 'import webarena' to 'import webarena_verified' and update all references.
affects: >=1.2.0
breakingv1.2.0 moved BrowserManager from 'webarena_verified.browser' to 'webarena_verified.utils'.
fix
Update import to 'from webarena_verified.utils import BrowserManager'.
affects: >=1.2.0
deprecatedThe 'llm_backend' parameter in WebArenaEnv is deprecated and will be removed in v2.0. Use 'agent' parameter instead.
fix
Replace 'llm_backend="gpt-4"' with 'agent="gpt-4"'.
affects: >=1.1.0 <=1.2.3
gotchaPlaywright browser binaries are not automatically installed. Missing browsers cause obscure errors.
fix
Run 'playwright install' after installing the package.
affects: *
Upgrade
Version history
1.2.3latest on PyPI · released Feb 7, 2026
Audit
Dependencies
playwrightrequiredRequired for browser automation in tests.
llm-interfaceoptionalUsed to interact with language model agents.
Agent activity
36 hits · last 30 days
node
32
OpenAI (training)
1
Resources

No resource links recorded.

webarena-verified — pip install webarena-verified · libregistry