Benchmark profile

3DCodeBench

An open benchmark for agentic procedural 3D modeling through generated code.

Data verified August 12, 202626 confirmed releases in the last 30 daysStart free brief

Benchmark score on 3DCodeBench — August 12, 2026

BenchLM mirrors the published score view for 3DCodeBench. Grok 4.6 leads the public snapshot at 54.0% , followed by Grok 4.5 (49.8%) and Claude Opus 4.8 (47.0%). We do not use these results to rank models overall.

1Closed

Grok 4.6

xAI

grok-4-6

54.0%

Overall 62.44Context 500K

2Closed

Grok 4.5

xAI

grok-4-5

49.8%

Overall 75.42Context 500K

3Closed

Claude Opus 4.8

Anthropic

claude-opus-4-8

47.0%

Overall 77.04Context 1M

5 modelsCodingCurrentDisplay onlyUpdated August 12, 2026

Benchmark score table (5 models)

Score

Grok 4.6xAI · Closed

54.0%

Grok 4.5xAI · Closed

49.8%

Claude Opus 4.8Anthropic · Closed

47.0%

Claude Fable 5Anthropic · Closed

43.7%

Claude Sonnet 5Anthropic · Closed

39.2%

The published 3DCodeBench snapshot places Grok 4.6 first at 54.0%. The third row is 7.0 points behind. The broader top-10 range is 14.8 points, so the table still separates the published systems.

5 models have been evaluated on 3DCodeBench. The benchmark falls in the Coding category. This category carries a 20% weight in BenchLM.ai's overall scoring system. 3DCodeBench is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About 3DCodeBench

Year

2026

Tasks

Procedural 3D modeling tasks

Format

Task score

Difficulty

Professional 3D modeling

The public site and GitHub repository expose the benchmark and result artifacts. BenchLM keeps the xAI comparison rows display-only because the card's setup is a specific agent run.

3DCodeBench leaderboard

BenchLM freshness & provenance

Version

3DCodeBench 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does 3DCodeBench measure?

An open benchmark for agentic procedural 3D modeling through generated code.

Which model scores highest on 3DCodeBench?

Grok 4.6 by xAI currently leads with a score of 54.0% on 3DCodeBench.

How many models are evaluated on 3DCodeBench?

5 AI models have been evaluated on 3DCodeBench on BenchLM.

Compare Top Models on 3DCodeBench

Grok 4.6 vs Grok 4.5 Grok 4.5 vs Claude Opus 4.8 Claude Opus 4.8 vs Claude Fable 5 Claude Fable 5 vs Claude Sonnet 5

Last updated: August 12, 2026 · BenchLM version 3DCodeBench 2026

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.