Atlas

Benchmarks

← All benchmarks

UnityBench-Mini3D

Code · 2026-10-26

77 project-level Unity tasks from Unity China's AI Hub across seven skill categories (scene construction, animation, particles and shaders, physics, user interface, editor operations and miscellaneous) and three difficulty tiers. A coding agent modifies a complete initial Unity project offline within one hour using its harness's file, shell and editor tools. After a clean reimport and compile, 1,178 held-out human-authored EditMode and PlayMode milestone tests are injected and run in the Unity Test Runner. A task scores the share of its milestone tests that pass, and the score is the unweighted mean over the 77 tasks, in percent.

Top models (higher is better)

ModelScore
GPT-5.6 Sol72.5
Claude Opus 571.8
Grok 4.571.2
Grok 4.670.6
GPT-5.6 Terra68.0
GPT-5.6 Luna66.4
Kimi K366.3
GPT-5.566.2
GLM-5.262.5
Sonnet 554.0
Kimi K2.7 Code41.5
Gemini 3.7 Flash31.2
Gemini 3.1 Pro Preview25.2
Loading Atlas data…