WebVR
Models recreate a webpage from a screen recording and supplied assets. A model judge scores the rendered page against visual and interaction criteria.
The authors rank 19 models on 175 webpages. They report separate scores for layout and motion.
- Made by
- StepFun and Tsinghua University
- Released
- March 2026
- Kind
- Web and app front ends
- Level
- University
- Scored
- Graded by a model
- Answers
- Code
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Strongest model tested
- Claude Opus 4.6 #39 on the capability index
- Last model added
- February 2026
- Scores refreshed
- November 20th, 2026
Kimi K2.5 leads with 79.1 among models released by February 2026
Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in February 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 79.1 Kimi K2.5
- 78.5 Claude Sonnet 4.6
- 77.9 GPT-5.2
- 77.3 Claude Opus 4.6
- 76.7
Gemini 3.1 Pro
- 71.9
Seed2.0 Pro
- 69.5
Gemini 3 Flash
- 67.3
Gemini 3 Pro
- 63.1
Gemini 2.5 Pro
- 62.0
Seed1.8
- 61.3
Qwen3.5 397B-A17B
- 61.2 Claude 3.7 Sonnet
- 55.6
Gemini 2.5 Flash
- 46.8
Qwen3-VL-235B-A22B
- 45.9 GPT-4.1
- 40.7
Qwen3-VL-235B-A22B-Instruct
- 37.7
Qwen3-VL-30B-A3B-Thinking
- 21.4
Qwen3-VL-30B-A3B-Instruct
- 11.4 GLM-4.6V
- 1 Kimi K2.5 79.1
- 2 Claude Sonnet 4.6 78.5
- 3 GPT-5.2 77.9
- 4 Claude Opus 4.6 77.3
- 5
Gemini 3.1 Pro 76.7
- 6
Seed2.0 Pro 71.9
- 7
Gemini 3 Flash 69.5
- 8
Gemini 3 Pro 67.3
- 9
Gemini 2.5 Pro Jun 2025 63.1
- 10
Seed1.8 62.0
- 11
Qwen3.5 397B-A17B 61.3
- 12 Claude 3.7 Sonnet 61.2
- 13
Gemini 2.5 Flash Sep 2025 55.6
- 14
Qwen3-VL-235B-A22B thinking 46.8
- 15 GPT-4.1 45.9
- 16
Qwen3-VL-235B-A22B-Instruct 40.7
- 17
Qwen3-VL-30B-A3B-Thinking 37.7
- 18
Qwen3-VL-30B-A3B-Instruct 21.4
- 19 GLM-4.6V 11.4
Source: Official WebVR leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Kimi K2.5 | 79.1 | Jan 2026 | Official leaderboard | |
| 2 | Claude Sonnet 4.6 | 78.5 | Feb 2026 | Official leaderboard | |
| 3 | GPT-5.2 | 77.9 | Dec 2025 | Official leaderboard | |
| 4 | Claude Opus 4.6 | 77.3 | Feb 2026 | Official leaderboard | |
| 5 | | 76.7 | Feb 2026 | Official leaderboard | |
| 6 | | 71.9 | Feb 2026 | Official leaderboard | |
| 7 | | 69.5 | Dec 2025 | Official leaderboard | |
| 8 | | 67.3 | Nov 2025 | Official leaderboard | |
| 9 | | 63.1 | Jun 2025 | Official leaderboard | |
| 10 | | 62.0 | Dec 2025 | Official leaderboard | |
| 11 | | 61.3 | Feb 2026 | Official leaderboard | |
| 12 | Claude 3.7 Sonnet | 61.2 | Feb 2025 | Official leaderboard | |
| 13 | | 55.6 | Sep 2025 | Official leaderboard | |
| 14 | | 46.8 | thinking | Sep 2025 | Official leaderboard |
| 15 | GPT-4.1 | 45.9 | Apr 2025 | Official leaderboard | |
| 16 | | 40.7 | Sep 2025 | Official leaderboard | |
| 17 | | 37.7 | Oct 2025 | Official leaderboard | |
| 18 | | 21.4 | Oct 2025 | Official leaderboard | |
| 19 | GLM-4.6V | 11.4 | Dec 2025 | Official leaderboard |
Rankings by part
We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Global Aesthetics, Navigation and Footer and Section-Specific Layouts. Kimi K2.5 leads 2 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Global Aesthetics
19 models we track have a score here.
- 1 GPT-5.2 ↑2 89.8
- 2
Gemini 3.1 Pro ↑3 88.3
- 3 Claude Opus 4.6 ↑1 87.7
- 4 Kimi K2.5 ↓3 87.4
- 5 Claude Sonnet 4.6 ↓3 87.2
- 6
Gemini 3 Flash ↑1 84.0
- 7
Seed2.0 Pro ↓1 82.9
- 8
Gemini 3 Pro 80.8
- 9
Qwen3.5 397B-A17B ↑2 80.5
- 10
Gemini 2.5 Pro Jun 2025 ↓1 78.6
Navigation and Footer
19 models we track have a score here.
- 1 Claude Sonnet 4.6 ↑1 89.4
- 2 Kimi K2.5 ↓1 89.2
- 3 GPT-5.2 89.1
- 4 Claude Opus 4.6 88.0
- 5
Gemini 3.1 Pro 87.3
- 6
Seed2.0 Pro 86.3
- 7
Gemini 3 Flash 85.2
- 8
Gemini 3 Pro 81.8
- 9 Claude 3.7 Sonnet ↑3 80.5
- 10
Gemini 2.5 Pro Jun 2025 ↓1 80.2
Section-Specific Layouts
19 models we track have a score here.
- 1 Kimi K2.5 79.3
- 2 Claude Sonnet 4.6 78.9
- 3 Claude Opus 4.6 ↑1 78.6
- 4 GPT-5.2 ↓1 77.3
- 5
Gemini 3.1 Pro 77.1
- 6
Seed2.0 Pro 73.3
- 7
Gemini 3 Flash 67.7
- 8
Gemini 3 Pro 66.3
- 9
Seed1.8 ↑1 62.2
- 10
Gemini 2.5 Pro Jun 2025 ↓1 59.6
Interaction and Motion
19 models we track have a score here.
- 1 Kimi K2.5 60.1
- 2 GPT-5.2 ↑1 60.0
- 3 Claude Sonnet 4.6 ↓1 59.1
- 4
Gemini 3.1 Pro ↑1 56.5
- 5 Claude Opus 4.6 ↓1 54.3
- 6
Gemini 2.5 Pro Jun 2025 ↑3 48.7
- 7
Gemini 3 Flash 48.4
- 8
Gemini 3 Pro 46.9
- 9
Seed2.0 Pro ↓3 45.9
- 10
Qwen3.5 397B-A17B ↑1 42.0
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official WebVR leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official WebVR leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Coding
- SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
- LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests.
DeepSeek-V4-Pro 93.5%
- SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
- DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
- SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases.
Seed2.1 Pro 59.8%
- Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance.
DeepSeek V4.1 Flash 3,471
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 9 months ago, in February 2026, for Gemini 3.1 Pro. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (19 entries).
- Pass Models scored 19 models we track.
- Warning Newest score The newest came 9 months ago, in February 2026, for Gemini 3.1 Pro.
- Note Frontier models None of the top 10 has a score. The strongest tested is Claude Opus 4.6 (#39).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on webvr-benchmark.github.io: 19 rows for models we track at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.