Benchmark Almanac

Vision2Web S1-2026

Claude Opus 4.5 52.7% leads since March 2026
Not in model cards

Agents build websites from 918 prototype images across 193 tasks, from static pages to full-stack apps. A GUI agent and a VLM judge score them.

This season uses Gemini-3-pro-preview as its visual judge. Scores from other seasons use different judge or agent configurations.

Made by
Z.ai (zai-org)
Released
March 2026
Kind
Web and app front ends
Level
Expert
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Strongest model tested
Gemini 3 Pro #55 on the capability index
Last model added
December 2025
Scores refreshed
November 12th, 2026

In the authors’ words

Recent advances in large language models have improved the capabilities of coding agents, yet systematic evaluation of complex, end-to-end website development remains limited. To address this gap, we introduce Vision2Web, a hierarchical benchmark for visual website development, spanning from static UI-to-code generation, interactive multi-page frontend reproduction, to long-horizon full-stack website development. The benchmark is constructed from real-world websites and comprises a total of 193 tasks across 16 categories, with 918 prototype images and 1,255 test cases. To support flexible, thorough and reliable evaluation, we propose workflow-based agent verification paradigm based on two complementary components: a GUI agent verifier and a VLM-based judge. We evaluate multiple visual language models instantiated under different coding-agent frameworks, revealing substantial performance gaps at all task levels, with state-of-the-art models still struggling on full-stack development.

Abstract of Vision2Web: A Hierarchical Benchmark for Visual Website Development with Agent Verification (arXiv:2603.26648, CC0)
8
Entries on its own leaderboard, 5 models we track

Claude Opus 4.5 leads with 52.7% among models released by December 2025

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

The newest model with a score here came out in December 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 52.7 Claude Opus 4.5
  2. 40.5 GPT-5
  3. 37.5 Claude Sonnet 4.5
  4. 36.1 Gemini 3 Pro
  5. 30.8 Gemini 3 Flash
  6. 9.40 Seed-1.8-VL
  7. 4.00 Qwen3-VL-8B
  8. 0.30 Qwen3-VL-32B
  1. 1 Claude Opus 4.5 52.7%
  2. 2 GPT-5 40.5%
  3. 3 Claude Sonnet 4.5 37.5%
  4. 4 Gemini 3 Pro 36.1%
  5. 5 Gemini 3 Flash 30.8%
  6. 6 Seed-1.8-VL 9.40%
  7. 7 Qwen3-VL-8B 4.00%
  8. 8 Qwen3-VL-32B 0.30%

Source: Official Vision2Web S1-2026 leaderboard

#EntryScoreSettingDateSource
1 Claude Opus 4.5 52.7% OpenHands Nov 2025 Official leaderboard
2 GPT-5 40.5% OpenHands Aug 2025 Official leaderboard
3 Claude Sonnet 4.5 37.5% OpenHands Sep 2025 Official leaderboard
4 Gemini 3 Pro 36.1% OpenHands Nov 2025 Official leaderboard
5 Gemini 3 Flash 30.8% OpenHands Dec 2025 Official leaderboard
6 Seed-1.8-VL 9.40% Claude Code Dec 2025 Official leaderboard
7 Qwen3-VL-8B 4.00% Claude Code Oct 2025 Official leaderboard
8 Qwen3-VL-32B 0.30% Claude Code Oct 2025 Official leaderboard

Number one over time

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

Anthropic
Own leaderboard Replaced by S2-2026, Mar 2026
Claude Opus 4.5 52.7%

Source: Official Vision2Web S1-2026 leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.5 GPT-5 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.5 GPT-5

Source: Official Vision2Web S1-2026 leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

Versions

We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.

Anthropic OpenAI
Vision2Web S3-2026 Not tested since Jun 2026
GPT-5.5 74.6%
Vision2Web S1-2026 Replaced by S2-2026, Mar 2026
Claude Opus 4.5 52.7%
Vision2Web S2-2026 Replaced by S3-2026, Jun 2026
Claude Opus 4.6 43.5%
Vision2Web S3-2026 Not tested since Jun 2026
GPT-5.5 74.6%

Source: Official leaderboards

Replaced by Vision2Web S2-2026

More in Coding

All 282 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