Benchmark Almanac

WebVR

Kimi K2.5 79.1 leads since March 2026
Not in model cards

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.

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
In and out
Video, Text, Images → Code
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
19
Models scored on its own leaderboard

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:

  1. 1 Kimi K2.5 79.1
  2. 2 Claude Sonnet 4.6 78.5
  3. 3 GPT-5.2 77.9
  4. 4 Claude Opus 4.6 77.3
  5. 5 Gemini 3.1 Pro 76.7
  6. 6 Seed2.0 Pro 71.9
  7. 7 Gemini 3 Flash 69.5
  8. 8 Gemini 3 Pro 67.3
  9. 9 Gemini 2.5 Pro Jun 2025 63.1
  10. 10 Seed1.8 62.0
  11. 11 Qwen3.5 397B-A17B 61.3
  12. 12 Claude 3.7 Sonnet 61.2
  13. 13 Gemini 2.5 Flash Sep 2025 55.6
  14. 14 Qwen3-VL-235B-A22B thinking 46.8
  15. 15 GPT-4.1 45.9
  16. 16 Qwen3-VL-235B-A22B-Instruct 40.7
  17. 17 Qwen3-VL-30B-A3B-Thinking 37.7
  18. 18 Qwen3-VL-30B-A3B-Instruct 21.4
  19. 19 GLM-4.6V 11.4

Source: Official WebVR leaderboard

#ModelScoreSettingDateSource
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 Gemini 3.1 Pro 76.7 Feb 2026 Official leaderboard
6 Seed2.0 Pro 71.9 Feb 2026 Official leaderboard
7 Gemini 3 Flash 69.5 Dec 2025 Official leaderboard
8 Gemini 3 Pro 67.3 Nov 2025 Official leaderboard
9 Gemini 2.5 Pro Jun 2025 63.1 Jun 2025 Official leaderboard
10 Seed1.8 62.0 Dec 2025 Official leaderboard
11 Qwen3.5 397B-A17B 61.3 Feb 2026 Official leaderboard
12 Claude 3.7 Sonnet 61.2 Feb 2025 Official leaderboard
13 Gemini 2.5 Flash Sep 2025 55.6 Sep 2025 Official leaderboard
14 Qwen3-VL-235B-A22B 46.8 thinking Sep 2025 Official leaderboard
15 GPT-4.1 45.9 Apr 2025 Official leaderboard
16 Qwen3-VL-235B-A22B-Instruct 40.7 Sep 2025 Official leaderboard
17 Qwen3-VL-30B-A3B-Thinking 37.7 Oct 2025 Official leaderboard
18 Qwen3-VL-30B-A3B-Instruct 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. 1 GPT-5.2 ↑2 89.8
  2. 2 Gemini 3.1 Pro ↑3 88.3
  3. 3 Claude Opus 4.6 ↑1 87.7
  4. 4 Kimi K2.5 ↓3 87.4
  5. 5 Claude Sonnet 4.6 ↓3 87.2
  6. 6 Gemini 3 Flash ↑1 84.0
  7. 7 Seed2.0 Pro ↓1 82.9
  8. 8 Gemini 3 Pro 80.8
  9. 9 Qwen3.5 397B-A17B ↑2 80.5
  10. 10 Gemini 2.5 Pro Jun 2025 ↓1 78.6

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Moonshot AI
Own leaderboard Not tested since Mar 2026
Kimi K2.5 79.1

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.

Plot with other benchmarks
On the benchmark’s own leaderboard
0.00 20.0 40.0 60.0 80.0 2025 2026 0.00 20.0 40.0 60.0 80.0 2025 2026

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

All 523 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.

How we check