WebForge-Bench - Consumer Transaction: leaderboard

WebForge-Bench (2026) tests browser agents on tasks set in automatically generated, reproducible websites with injected real-web noise, controlled for difficulty along seven dimensions. This board is task accuracy in percent on the consumer transaction and service domain with screenshot and DOM input; higher is better.

Metric: Accuracy (%) on the consumer transaction and service tasks of the 934 WebForge-Bench tasks on generated, self-contained websites with injected real-web noise, up to 50 browser actions in a Chromium GUI, final-state answer checking; multimodal setting: screenshot plus DOM; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 10 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro72.2#145
2Gemini 3 Flash65.2#175
3Kimi K2.560#266
4Claude Sonnet 4.558.3#261
5GPT-5 Mini51.3#364
6GPT-5.248.7#185
7Qwen 3 VL 235B A22B37.4#485
8Gemini 2.5 Flash Lite34.8#964
9GPT-5 Nano20.9#906
10qwen3-omni-30B6.1#1608

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=webforge-bench-consumer-transaction · How It Works · Data refreshed daily, snapshot 2026-11-21.