WebForge-Bench - Tool Usage: 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 tool usage domain with screenshot and DOM input; higher is better.

Metric: Accuracy (%) on the tool usage 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: Estimated already saturated. 10 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro76.6#145
2Claude Sonnet 4.567.4#261
3Gemini 3 Flash66#175
4GPT-5.263.1#185
5Kimi K2.561.7#266
6GPT-5 Mini58.2#364
7Qwen 3 VL 235B A22B48.2#485
8Gemini 2.5 Flash Lite33.3#964
9GPT-5 Nano29.8#906
10qwen3-omni-30B12.1#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-tool-usage · How It Works · Data refreshed daily, snapshot 2026-11-21.