This benchmark evaluates the 3D understanding of language models: each model writes three.js code that builds an object from a short text prompt, and the object it builds is what gets judged. The rankings come from 365,552 pairwise matches, decided by 1,932,427 responses. Quality is evaluated in four dimensions: Alignment, Coherence, Creativity and Preference.We wrote 500 text prompts in five categories of 100: vehicles, animals and creatures, buildings and architecture, furniture and household objects, and instruments and tools. All models share a ceiling of 64,000 output tokens per call, thinking included. Only ERNIE 5.5 Flash Preview reaches it often, in 243 of its 690 calls, and every model's median call stays below 14,000 tokens except four: ERNIE 5.5 Flash Preview at 57,143, Gemini 4 Carbon at 42,463, DeepSeek V4.1 Flash at 22,501 and Qwen4 Max at 18,290 (see the output and cost table).See the details for more and the methodology for evaluation and scoring.
We use crowd intelligence to deliver real, diverse human feedback at scale for model evaluation and post-training.