Omni2Web - Instruction Recovery Target F1: leaderboard
The target component of Omni2Web instruction recovery (Han et al., 2026): F1 in percent of the recovered edit targets against the reference targets, from 918 bilingual screen recordings of weakly referential web-editing requests, where speech, cursor movement and edit history identify the targets. Extra or duplicate predictions lower the score; higher is better.
Metric: Target F1 (%; Track B component: recovered edit targets matched one-to-one to the reference targets; 918 bilingual screen recordings of weakly referential webpage edit requests; the model rewrites the recording into an ordered list of explicit target and action instructions without the source HTML; one-to-one matching against reference instructions, semantic equivalence judged by Gemini 3.1 Pro; unusable outputs score zero). Source: arxiv.org. Saturation forecast: Around 2029. 17 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 Omni Plus | 49.99 | #237 |
| 2 | Claude Opus 5 | 48.06 | #20 |
| 3 | Seed 2.0 Lite | 47.91 | #286 |
| 4 | GPT-5.6 Sol | 47.51 | #29 |
| 5 | Gemini 3.5 Flash | 47.04 | #98 |
| 6 | Gemini 3.1 Pro (Preview) | 42.19 | #91 |
| 7 | Kimi K3 | 38.64 | #53 |
| 8 | Qwen 3.8 Max | 34.84 | #55 |
| 9 | Grok 4.6 | 34.6 | #44 |
| 10 | MiMo-V2.5 | 33.95 | #353 |
| 11 | Seed 2.1 Pro | 33.21 | #133 |
| 12 | MiniMax-M3 | 21.25 | #265 |
| 13 | Qwen 3.8 27B | 19.82 | #199 |
| 14 | Muse Spark 1.1 | 18.45 | #82 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=omni2web-instruction-recovery-target-f1 · How It Works · Data refreshed daily, snapshot 2026-11-12.