Benchmark Almanac

3D SubwayBench

Claude Opus 5 5.00 leads since July 2026
Not in model cards

Agents use Command Code to build a Subway Surfers style endless runner in a single HTML file. The scorecard grades six features from the submitted code.

Command Code scores six models. The reviewers inspected the code without playing the games. The rubric gives no points for saving a high score.

Made by
Command Code
Released
July 2026
Kind
Building whole projects
Level
University
Scored
Graded by people
Answers
Code
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Code, Actions
Interaction
Agentic
Environment
Terminal
Items from
Written by experts
Refresh
One fixed set
Run by
Anyone
In views
Agents
Strongest model tested
Claude Opus 5 #11 on the capability index
Last model added
July 2026
Scores refreshed
November 20th, 2026
6
Models scored on its own leaderboard

Claude Opus 5 leads with 5.00 among models released by July 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 July 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 Claude Opus 5 5.00
  2. 2 GPT-5.6 Sol 4.75
  3. 3 Grok 4.5 4.50
  4. 4 Claude Fable 5 4.33
  5. 5 GPT-5.5 4.08
  6. 6 Gemini 3.6 Flash 3.50

Source: Official 3D SubwayBench leaderboard

#ModelScoreSettingDateSource
1 Claude Opus 5 5.00 Jul 2026 Official leaderboard
2 GPT-5.6 Sol 4.75 Jul 2026 Official leaderboard
3 Grok 4.5 4.50 Jul 2026 Official leaderboard
4 Claude Fable 5 4.33 Jun 2026 Official leaderboard
5 GPT-5.5 4.08 Apr 2026 Official leaderboard
6 Gemini 3.6 Flash 3.50 Jul 2026 Official leaderboard

Rankings by part

We read 6 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Spec fidelity, Collision and Architecture. GPT-5.5 leads 2 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Spec fidelity

6 models we track have a score here.

  1. 1 GPT-5.5 ↑4 5.00
  2. 2 Claude Fable 5 ↑2 5.00
  3. 3 Grok 4.5 5.00
  4. 4 GPT-5.6 Sol ↓2 5.00
  5. 5 Claude Opus 5 ↓4 5.00
  6. 6 Gemini 3.6 Flash 2.50

4 models have held number one on the official leaderboard since its release in July 2026

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

Anthropic xAI OpenAI
Own leaderboard Not tested since Jul 2026

Source: Official 3D SubwayBench 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
3.50 4.00 4.50 5.00 2026 3.50 4.00 4.50 5.00 2026

Source: Official 3D SubwayBench 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 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (6 entries).
  • Pass Models scored 6 models we track.
  • Pass Newest score The newest is from July 2026, for Claude Opus 5.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Claude Opus 5 (#11).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 6 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