AgentWorldBench Web

AgentWorldBench Web is a benchmark for how well a model acts as a general-purpose agent. The official description says: "AgentWorldBench Web domain - open-ended rubric evaluation for language world models simulating web browser environments". Below we describe in simple words what benchmarks of this type generally examine.

AgentWorldBench Web measures how well a model acts as a general-purpose agent. In detail, it checks:

• solving long, multi-step tasks with many intermediate decisions
• using tools correctly (browsers, terminals, APIs, calendars, other apps)
• adapting when something unexpected happens mid-task
• staying on goal over long task horizons instead of wandering off

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.

Official methodology note: AgentWorldBench Web domain - open-ended rubric evaluation for language world models simulating web browser environments. Normalized to 0-100.

What is the AgentWorldBench Web benchmark?

AgentWorldBench Web is a benchmark for how well a model acts as a general-purpose agent. The official description says: 'AgentWorldBench Web domain - open-ended rubric evaluation for language world models simulating web browser environments'. Below we describe in simple words what benchmarks of this type generally examine.

What does AgentWorldBench Web test?

AgentWorldBench Web mainly measures how well a model acts as a general-purpose agent. Concretely, it checks: solving long, multi-step tasks with many intermediate decisions; using tools correctly (browsers, terminals, APIs, calendars, other apps); adapting when something unexpected happens mid-task.

How is the AgentWorldBench Web score calculated?

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better. The declared maximum score of this benchmark is 100.0.

Where do I get AgentWorldBench Web data and scores?

The official source is https://huggingface.co/Qwen/Qwen-AgentWorld-35B-A3B. The official benchmark page/reference is linked on this page. Every score on Open LLM Radar is also available as open data: download the scores JSON (http://open-llm-radar-api:8000/api/v1/benchmarks/agentworldbench-web/scores) or pull it directly from the Open LLM Radar API with any HTTP client.