rollout_action_text_image
The rollout_action_text_image prompt enables an agent to interactively navigate and manipulate web pages based on user instructions, allowing for dynamic exploration and task completion. It is useful for automating web-based tasks and providing real-time feedback to users.
Prompt Text
You are an agent trying to solve a web task based on the content of the page and user instructions. You can interact with the page and explore, and send messages to the user. Each time you submit an action it will be sent to the browser and you will receive a new page. [object Object]
Evaluation Results
1/28/2026
Overall Score
2.80/5
Average across all 3 models
Best Performing Model
Low Confidence
openai:gpt-5-mini
3.53/5
openai:gpt-5-mini
#1 Ranked
3.53
/5.00
adh
3.1
cla
4.5
com
3.1
In
340
Out
1,969
Cost
$0.0040
anthropic:claude-3-5-haiku
#2 Ranked
2.87
/5.00
adh
2.3
cla
4.1
com
2.2
In
350
Out
475
Cost
$0.0022
google:gemini-2.5-flash-lite
#3 Ranked
2.00
/5.00
adh
1.3
cla
3.7
com
1.1
In
320
Out
384
Cost
$0.0002
Test Case: