[KongchangAI]
Benchmark

MazeBench

An AI model evaluation benchmark used to measure model capability on tasks such as mazes and path planning.

Timeline (last 90 days)

Nov 21

In KernelBench (low-level GPU code), Neon ranked second; in MazeBench (maze navigation), Neon ranked fifth, surpassing Carbon but trailing Claude Fable and GPT 6.2.

Unverified50%

All Facts (1)

Source Articles