Vision2Web
We show this table for reference; we do not rank on it.
A benchmark for converting visual references into functional web implementations.
Benchmark score on Vision2Web — November 13, 2026
We compile the Vision2Web rows from provider self-reports. LongCat-2.5 leads the table at 78.6%, followed by Qwen3.8 Max (69.0%) and Qwen3.8-Flash-Next (64.0%). We do not use these results to rank models overall.
LongCat-2.5
Meituan
Reported by MeituanQwen3.8 Max
Alibaba
Reported by AlibabaQwen3.8-Flash-Next
Alibaba
Reported by Alibaba5 modelsMultimodal & GroundedCurrentDisplay onlyUpdated November 13, 2026
Benchmark score table (5 models)
All 5 results on this page were reported by the labs that built the models.
| Rank | Model / configuration | Score | Who ran it | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | LongCat-2.5MeituanReported by Meituan | 78.6% | Reported by Meituan | Not reported | Open |
| 2 | Qwen3.8 MaxAlibabaReported by Alibaba | 69.0% | Reported by Alibaba | Not reported | Open |
| 3 | Qwen3.8-Flash-NextAlibabaReported by Alibaba | 64.0% | Reported by Alibaba | Not reported | Open |
| 4 | Qwen3.8-27BAlibabaReported by Alibaba | 62.9% | Reported by Alibaba | Not reported | Open |
| 5 | Qwen3.8-Omni-FlashAlibabaReported by Alibaba | 62.9% | Reported by Alibaba | Not reported | Closed |
- Lab-reported
- Benchmark-run
- Independent
Among the reported Vision2Web rows, LongCat-2.5 is first at 78.6%. The third row is 14.6 points behind. The broader top-10 range is 15.7 points, so the table still separates the published systems.
5 models have been evaluated on Vision2Web. The benchmark falls in the Multimodal & Grounded category. Vision2Web is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About Vision2Web
Year
2026
Tasks
Screenshot-to-web tasks
Format
Visual reference to web implementation
Difficulty
Multimodal web generation
BenchLM stores Vision2Web as a display-only screenshot-to-web benchmark reference outside the weighted core schema.
Freshness and provenance
Version
Vision2Web 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does Vision2Web measure?
A benchmark for converting visual references into functional web implementations.
Which model scores highest on Vision2Web?
LongCat-2.5 by Meituan currently leads with a score of 78.6% on Vision2Web.
How many models are evaluated on Vision2Web?
5 AI models have published results on Vision2Web in the BenchLM catalog.
Compare top models on Vision2Web
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.