WebArena
by CMU · free · Last verified 2026-03-01
WebArena is a realistic and reproducible benchmark environment designed to evaluate autonomous language agents. It tests an agent's ability to perform complex, multi-step tasks across a diverse set of self-hosted websites, including e-commerce, forums, and content management systems, using real web interfaces.
https://webarena.dev ↗B
B—Above Average
Adoption: BQuality: A+Freshness: ACitations: B+Engagement: F
Specifications
- License
- Apache-2.0
- Pricing
- free
- Capabilities
- Autonomous Agent Evaluation, Complex Task Completion Benchmarking, Natural Language Instruction Following, Reproducible Web Environment Testing, Cross-Domain Web Interaction, Information Retrieval and Synthesis, Form Filling and User Input Simulation, Performance Measurement on Realistic Websites
- Integrations
- Use Cases
- [object Object], [object Object], [object Object], [object Object], [object Object]
- API Available
- No
- Evaluated Models
- claude-4, gpt-5, gemini-2.5-pro, deepseek-v3
- Metrics
- success-rate, step-accuracy
- Methodology
- 812 web-based tasks across 5 self-hosted websites. Agents interact via browser actions and are evaluated on task completion determined by URL, page content, or database state checks.
- Last Run
- 2026-02-28
- Tags
- benchmark, agent-evaluation, web-benchmark, autonomous-agents, browser-automation, llm-evaluation, reproducible-research, web-environment, reinforcement-learning, human-computer-interaction
- Added
- 2026-03-17
- Completeness
- 0.9%
Index Score
62.4Adoption
66
Quality
90
Freshness
86
Citations
72
Engagement
0