Vision2Web S3-2026
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-3.5-flash as its visual judge. Scores from other seasons use different judge or agent configurations.
- Made by
- Z.ai (zai-org)
- Released
- June 2026
- Kind
- Web and app front ends
- Level
- Expert
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Strongest model tested
- GPT-5.5 #20 on the capability index
- Last model added
- June 2026
- 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.
GPT-5.5 leads with 74.6% among models released by June 2026
Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in June 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- Claude Fable 5.2
-
Gemini 4 Carbon - GPT-6 Astra
- Claude Opus 5.5
- and 15 more
- 1 GPT-5.5 74.6%
- 2 Claude Opus 4.8 71.4%
- 3 Claude Opus 4.7 65.9%
- 4
MiniMax-M3 59% - 5 Claude Sonnet 4.6 58.9%
- 6 Claude Opus 4.6 57.6%
- 7 Kimi K2.6 54.8%
Source: Official Vision2Web S3-2026 leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.5 | 74.6% | Codex | Apr 2026 | Official leaderboard |
| 2 | Claude Opus 4.8 | 71.4% | Claude Code | May 2026 | Official leaderboard |
| 3 | Claude Opus 4.7 | 65.9% | Claude Code | Apr 2026 | Official leaderboard |
| 4 | MiniMax-M3 | 59% | Claude Code | Jun 2026 | Official leaderboard |
| 5 | Claude Sonnet 4.6 | 58.9% | Claude Code | Feb 2026 | Official leaderboard |
| 6 | Claude Opus 4.6 | 57.6% | Claude Code | Feb 2026 | Official leaderboard |
| 7 | Kimi K2.6 | 54.8% | Claude Code | Apr 2026 | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Vision2Web S3-2026 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 Vision2Web S3-2026 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.
Source: Official leaderboards
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