Benchmark Almanac

Frontier Design Benchmark

Gestalt 1,931 tops its leaderboard Claude Fable 5.2 1,713 leads the models we track
Not in model cards

Professional designers vote blind between two models’ designs for the same real client brief from Contra’s marketplace, such as a marketing site. Models are ranked by Elo from the votes.

Launched November 18, 2026 with 33 briefs, 68 designers and 4,311 votes. Each brief has a golden design by a Contra designer, which places second. A 35-dimension design rubric and an AI slop checklist add scores that correlate with the votes (r = 0.94). A stealth model, Gestalt, leads.

Made by
Contra Labs and BenchFlow
Released
November 2026
Kind
Web and app front ends
Level
Expert
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Measures
Preference
Subject
Arts
Language
English
In and out
Text → Images
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
A host
Occupations
Arts, design, entertainment, sports and media
Strongest model tested
Claude Fable 5.2 #2 on the capability index
Last model added
November 2026
Scores refreshed
November 18th, 2026
16
Entries on its own leaderboard, 15 models we track

Gestalt tops the leaderboard with 1,931 Claude Fable 5.2 leads with 1,713

Best score per entry from the benchmark's own leaderboard, read on November 18th, 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 best model we track is Claude Fable 5.2, at 1,713. This view shows only models we track, each at its best.

Bars start at 1,000.

  1. 1 Gestalt 1,931
  2. 2 Claude Fable 5.2 1,713
  3. 3 Claude Opus 5.5 1,691
  4. 4 GPT-6.1 Sol 1,668
  5. 5 Claude Sonnet 5.5 1,667
  6. 6 Kimi K3.1 Pro 1,663
  7. 7 GPT-6 Astra 1,638
  8. 8 Qwen 3.8 Max 1,434
  9. 9 Kimi K3 1,415
  10. 10 GLM-5.3 prime 1,403
  11. 11 Gemini 4 Carbon 1,366
  12. 12 Muse Spark 1.3 1,335
  13. 13 Muse Wonder 1,249
  14. 14 Grok 4.7 1,247
  15. 15 Gemini 4 Argon 1,221
  16. 16 DeepSeek-V4-Pro 1,121

Source: Official Frontier Design Benchmark leaderboard

#EntryScore ±SettingDateSource
1 Gestalt 1,931 ±72 Official leaderboard
2 Claude Fable 5.2 1,713 ±35 Nov 2026 Official leaderboard
3 Claude Opus 5.5 1,691 ±42 Sep 2026 Official leaderboard
4 GPT-6.1 Sol 1,668 ±43 Sep 2026 Official leaderboard
5 Claude Sonnet 5.5 1,667 ±39 Sep 2026 Official leaderboard
6 Kimi K3.1 Pro 1,663 ±40 Oct 2026 Official leaderboard
7 GPT-6 Astra 1,638 ±47 Sep 2026 Official leaderboard
8 Qwen 3.8 Max 1,434 ±45 Aug 2026 Official leaderboard
9 Kimi K3 1,415 ±47 Jul 2026 Official leaderboard
10 GLM-5.3 1,403 ±50 prime Aug 2026 Official leaderboard
11 Gemini 4 Carbon 1,366 ±42 Oct 2026 Official leaderboard
12 Muse Spark 1.3 1,335 ±45 Sep 2026 Official leaderboard
13 Muse Wonder 1,249 ±42 Oct 2026 Official leaderboard
14 Grok 4.7 1,247 ±42 Sep 2026 Official leaderboard
15 Gemini 4 Argon 1,221 ±47 Sep 2026 Official leaderboard
16 DeepSeek-V4-Pro 1,121 ±69 Apr 2026 Official leaderboard
#ModelScore ±SettingDateSource
1 Claude Fable 5.2 1,713 ±35 Nov 2026 Official leaderboard
2 Claude Opus 5.5 1,691 ±42 Sep 2026 Official leaderboard
3 GPT-6.1 Sol 1,668 ±43 Sep 2026 Official leaderboard
4 Claude Sonnet 5.5 1,667 ±39 Sep 2026 Official leaderboard
5 Kimi K3.1 Pro 1,663 ±40 Oct 2026 Official leaderboard
6 GPT-6 Astra 1,638 ±47 Sep 2026 Official leaderboard
7 Qwen 3.8 Max 1,434 ±45 Aug 2026 Official leaderboard
8 Kimi K3 1,415 ±47 Jul 2026 Official leaderboard
9 GLM-5.3 1,403 ±50 prime Aug 2026 Official leaderboard
10 Gemini 4 Carbon 1,366 ±42 Oct 2026 Official leaderboard
11 Muse Spark 1.3 1,335 ±45 Sep 2026 Official leaderboard
12 Muse Wonder 1,249 ±42 Oct 2026 Official leaderboard
13 Grok 4.7 1,247 ±42 Sep 2026 Official leaderboard
14 Gemini 4 Argon 1,221 ±47 Sep 2026 Official leaderboard
15 DeepSeek-V4-Pro 1,121 ±69 Apr 2026 Official leaderboard

Other ways to score it

The benchmark’s own leaderboard also reports Win rate, another way to score the whole test. Labs sometimes quote it instead of the headline. Claude Opus 5.5 leads it. Arrows show how far a model moved from its overall rank among the models we track.

The share of a model’s head-to-head votes it won. 15 models we track have a score here.

  1. 1 Claude Opus 5.5 ↑1 73%
  2. 2 Claude Fable 5.2 ↓1 73%
  3. 3 Kimi K3.1 Pro ↑2 71%
  4. 4 GPT-6.1 Sol ↓1 70%
  5. 5 Claude Sonnet 5.5 ↓1 69%
  6. 6 GPT-6 Astra 66%
  7. 7 GLM-5.3 ↑2 41%
  8. 8 Kimi K3 40%
  9. 9 Qwen 3.8 Max ↓2 39%
  10. 10 Gemini 4 Carbon 36%

2 models have held number one on the official leaderboard since its release in November 2026

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

Anthropic
Own leaderboard 2 leaders

Source: Official Frontier Design Benchmark 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
1,000 1,200 1,400 1,600 1,800 2026 2027 1,000 1,200 1,400 1,600 1,800 2026 2027

Source: Official Frontier Design Benchmark 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

All 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (16 entries).
  • Pass Models scored 15 models we track, plus 1 other entry on its leaderboard.
  • Pass Newest score The newest is from November 2026, for Claude Fable 5.2.
  • Pass Frontier models 7 of the top 10 have a score, led by Claude Fable 5.2 (#2).
  • Note Headroom Scored as an Elo rating, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on contralabs.com: 15 rows for models we track and 1 other entry at the last read.
  • Pass Last read November 18th, 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