agents benchmark
Web Bench
Web Bench evaluates agents on realistic web-development engineering tasks, measuring end-to-end implementation in browser-based workflows.
Updated Aug 7, 2026
Published models2
Registry coverage2
MetricScore
EvidenceB
Web Bench leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | BY | 78.4% | 100.0% | 2 | C | |
| 2 | BY | 73.6% | 0.0% | 2 | C |
Score distribution
Top published rows on the benchmark's original scale.
Web Bench highlights
The top published results on this benchmark's own scale.
What is Web Bench?
Definition and scoring fields from the benchmark registry.
Web Bench evaluates agents on realistic web-development engineering tasks, measuring end-to-end implementation in browser-based workflows.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- Web Bench
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- web-bench|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about Web Bench.
Which model scores highest on Web Bench?
Seed 2.1 Pro is currently ranked first with 78.4%.
What does Web Bench measure?
Web Bench evaluates agents on realistic web-development engineering tasks, measuring end-to-end implementation in browser-based workflows.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.