WebForge-Bench - Content Creation: 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 content creation and publishing domain with screenshot and DOM input; higher is better.

Metric: Accuracy (%) on the content creation and publishing 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 Pro80.9#146
2Claude Sonnet 4.576.3#267
3Gemini 3 Flash74.8#180
4Kimi K2.574.8#272
5GPT-5.271#190
6GPT-5 Mini67.9#373
7Qwen 3 VL 235B A22B51.1#486
8Gemini 2.5 Flash Lite39.7#972
9GPT-5 Nano30.5#912
10qwen3-omni-30B13#1612

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=webforge-bench-content-creation · How It Works · Data refreshed daily, snapshot 2026-11-22.