WebForge-Bench - Level 1: 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 difficulty level 1 with screenshot and DOM input; higher is better.
Metric: Accuracy (%) on the difficulty level 1 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
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro | 86.4 | #146 |
| 2 | Claude Sonnet 4.5 | 85.7 | #267 |
| 3 | Kimi K2.5 | 84.4 | #272 |
| 4 | Gemini 3 Flash | 82.4 | #180 |
| 5 | GPT-5 Mini | 82.4 | #373 |
| 6 | GPT-5.2 | 80.1 | #190 |
| 7 | Qwen 3 VL 235B A22B | 73.4 | #486 |
| 8 | GPT-5 Nano | 61.8 | #912 |
| 9 | Gemini 2.5 Flash Lite | 58.5 | #972 |
| 10 | qwen3-omni-30B | 26.9 | #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-level-1 · How It Works · Data refreshed daily, snapshot 2026-11-22.