Model Bake-Off for Code-Built 3D Assets

Evaluating reasoning models against specialized subagents for code-synthesized 3D environments.

Core Finding:When generating procedural 3D environments and assets directly via code, how do general high-effort reasoning models compare to domain-specific agent pipelines? In a blind-judged bake-off across multiple model families, general reasoning combined with visual reference loops decisively outperformed specialized builders.

Codex (gpt-6-astra)

58 / 80

Highest overall score; sole entry to generate photoreal procedural sky, water, and sand shaders.

Fable (Claude Subagent)

29 / 80

Specialized domain agent; struggled with complex shader graph compilation and lighting balance.

Gemini 3.8 Flash

27 / 80

Fast generator; produced basic geometric meshes but failed to achieve photoreal material properties.

Oracle Independence

100% Locked

Zero self-judging allowed; implementing agents prohibited from altering visual acceptance thresholds.

Empirical Findings Table

Blind evaluation scores in procedural sunset scene recreation

Telemetry Metric	Measured Value	Context & Observability
Codex (gpt-6-astra)	58 / 80	Highest overall score; sole entry to generate photoreal procedural sky, water, and sand shaders.
Fable (Claude Subagent)	29 / 80	Specialized domain agent; struggled with complex shader graph compilation and lighting balance.
Gemini 3.8 Flash	27 / 80	Fast generator; produced basic geometric meshes but failed to achieve photoreal material properties.
Oracle Independence	100% Locked	Zero self-judging allowed; implementing agents prohibited from altering visual acceptance thresholds.

Engineering Action Directives

Iterative visual comparison loops combined with high-effort reasoning models appear more effective for procedural shader synthesis than specialized domain prompts without visual feedback.
Enforcing strict external visual reference targets prevents models from shifting goals when generation artifacts occur.

Methodology Boundaries & Limits

Single visual target: Evaluated on a procedural coastal sunset environment in Blender; may not generalize to character modeling or interior architectural scenes.
Qualitative rubric: While scored double-blind on an 80-point rubric, visual quality evaluations carry inherent aesthetic judging variance.