Omni2Web - Direct Editing Fulfillment: leaderboard
The edit-fulfilment component of Omni2Web direct editing (Han et al., 2026): the share of rubric items whose requested change is actually realised in the edited HTML, over 918 bilingual screen recordings of weakly referential web-editing requests, where speech, cursor movement and edit history identify the targets. Percent, judged by Gemini 3.1 Pro; higher is better.
Metric: Edit-fulfillment rate (%; Track A rubric items: each active step's requested change is realised in the edited HTML; 918 bilingual screen recordings of weakly referential webpage edit requests (speech plus cursor), 13,907 edit steps; the model edits the source HTML directly from the recording under its native input interface (omni: audio-visual video; vision-language: video or 1 fps frames plus manual transcript); rubric items judged by Gemini 3.1 Pro; unusable outputs score zero). Source: arxiv.org. Saturation forecast: Around July 2027. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.5 Flash | 40.52 | #98 |
| 2 | Claude Opus 5 | 38.9 | #20 |
| 3 | GPT-5.6 Sol | 37.32 | #29 |
| 4 | Kimi K3 | 30.26 | #53 |
| 5 | Grok 4.6 | 29.92 | #44 |
| 6 | Qwen 3.8 Max | 27.59 | #55 |
| 7 | Seed 2.1 Pro | 23.67 | #133 |
| 8 | MiniMax-M3 | 18.83 | #265 |
| 9 | MiMo-V2.5 | 17.93 | #353 |
| 10 | Gemini 3.1 Pro (Preview) | 16.46 | #91 |
| 11 | Qwen 3.5 Omni Plus | 11.86 | #237 |
| 12 | Seed 2.0 Lite | 11.7 | #286 |
| 13 | Muse Spark 1.1 | 10.14 | #82 |
| 14 | Qwen 3.8 27B | 7.56 | #199 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=omni2web-direct-editing-fulfillment · How It Works · Data refreshed daily, snapshot 2026-11-12.