Benchmark Almanac

Floorplan Arena

Gemini 4 Carbon 1,792 leads since October 2026
Not in model cards

Models furnish hand-drawn room sketches. People compare the layouts and vote for the better floorplan.

The official API publishes Arena ratings, confidence intervals and vote counts.

Made by
Kaggle
Kind
Data science and SQL
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Strongest model tested
Gemini 4 Carbon #2 on the capability index
Last model added
October 2026
Scores refreshed
November 12th, 2026
11
Models scored on its own leaderboard

Gemini 4 Carbon leads with 1,792

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.

Bars start at 1,200.

  1. 1,792 Gemini 4 Carbon
  2. 1,742 Gemini 4 Argon
  3. 1,713 Claude Fable 5.1
  4. 1,608 GPT-6 Astra
  5. 1,488 Claude Opus 5.5
  6. 1,482 Gemini 3.8 Flash
  7. 1,473 GPT-6.1 Sol
  8. 1,403 GPT-6.1 Terra
  9. 1,380 GPT-6 Luna
  10. 1,270 Grok 4.7
  11. 1,257 Claude Sonnet 5.5

Bars start at 1,200.

  1. 1 Gemini 4 Carbon Gemini 4 Carbon High Reasoning 1,792
  2. 2 Gemini 4 Argon Gemini 4 Argon High Reasoning 1,742
  3. 3 Claude Fable 5.1 Claude Fable 5.1 Max Reasoning 1,713
  4. 4 GPT-6 Astra GPT-6 Astra Max Reasoning 1,608
  5. 5 Claude Opus 5.5 Claude Opus 5.5 Xhigh Reasoning 1,488
  6. 6 Gemini 3.8 Flash Gemini 3.8 Flash High Reasoning 1,482
  7. 7 GPT-6.1 Sol GPT-6.1 Sol Max Reasoning 1,473
  8. 8 GPT-6.1 Terra GPT-6.1 Terra Max Reasoning 1,403
  9. 9 GPT-6 Luna GPT-6 Luna Max Reasoning 1,380
  10. 10 Grok 4.7 Grok 4.7 Xhigh Reasoning 1,270
  11. 11 Claude Sonnet 5.5 Claude Sonnet 5.5 Xhigh Reasoning 1,257

Source: Official Floorplan Arena leaderboard

#ModelScoreSettingDateSource
1 Gemini 4 Carbon 1,792 Gemini 4 Carbon High Reasoning Oct 2026 Official leaderboard
2 Gemini 4 Argon 1,742 Gemini 4 Argon High Reasoning Sep 2026 Official leaderboard
3 Claude Fable 5.1 1,713 Claude Fable 5.1 Max Reasoning Sep 2026 Official leaderboard
4 GPT-6 Astra 1,608 GPT-6 Astra Max Reasoning Sep 2026 Official leaderboard
5 Claude Opus 5.5 1,488 Claude Opus 5.5 Xhigh Reasoning Sep 2026 Official leaderboard
6 Gemini 3.8 Flash 1,482 Gemini 3.8 Flash High Reasoning Sep 2026 Official leaderboard
7 GPT-6.1 Sol 1,473 GPT-6.1 Sol Max Reasoning Sep 2026 Official leaderboard
8 GPT-6.1 Terra 1,403 GPT-6.1 Terra Max Reasoning Oct 2026 Official leaderboard
9 GPT-6 Luna 1,380 GPT-6 Luna Max Reasoning Sep 2026 Official leaderboard
10 Grok 4.7 1,270 Grok 4.7 Xhigh Reasoning Sep 2026 Official leaderboard
11 Claude Sonnet 5.5 1,257 Claude Sonnet 5.5 Xhigh Reasoning Sep 2026 Official leaderboard

3 models have held number one on the official leaderboard since 2023

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

Anthropic Google DeepMind
Own leaderboard 3 leaders
Gemini 4 Carbon 1,792

Source: Official Floorplan Arena 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
1,200 1,400 1,600 1,800 2027 1,200 1,400 1,600 1,800 2027

Source: Official Floorplan Arena 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.

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