AI Bowser Comparison

A cost and performance benchmark of 3D character generation across frontier AI models.

Playable

Description

A comparison of 3D modeling capabilities between DeepSeek V4.1 Flash and Claude Opus 5.5. The project showcases a high-fidelity Bowser model generated with DeepSeek for $0.15 (8.05M tokens, 32m 42s) versus a version from Opus 5.5 costing $3.85 (3.88M tokens, 23m 42s), highlighting significant discrepancies in character accuracy and value.

Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.

Details

Why this is interesting: Impressive one-shot creation of a functional 3D game with physics and dialogue based on a specific viral video.
Attributed to: @noclipepe
Built with: Opus 5.5 GPT-6 Astra Gemini 4 Argon Fable 5.1 V4.1 Flash
Links: claude.ai
View tweet
Tags: #3d-modeling #character-generation #benchmark #gamedev

Discovered: X
First posted: X · Oct 6, 2026
Listed: Added 11d ago
Build: Strong
Models: Strong

Prompt

build a liquid-glass welcome screen in each model’s own theme.

Source