Benchmark RadarAI BENCHMARK PROFILE
WebIGBench
WebIGBench evaluates code generation for interactive webpages with 103 complex examples and 871 distinct actions, proposing an automated evaluation pipeline for interactive consistency.
- Released
- 2026-05-29
- Readiness
- Runnable
- Primary field
- General AI
Why it matters
Fills the gap in benchmarking interactive webpage code generation, offering a public dataset and evaluation method for model comparison.
Motivation
Recent advancements in multimodal large language models (MLLMs) have achieved remarkable progress in multimodal reasoning and code generation, catalyzing a new paradigm for front-end development.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.