Skip to benchmark content
MotionBenchHosted by Baz Studio
Pilot datasetOne prompt · one trial · 33 routes · not the official Core rankingRead the limits
openai · O+C

GPT 5.4 Mini

gpt-5.4-mini
Pilot visual tierB
Human assessment
Simple line-driven transition with a readable finish.

One diagnostic trial. This is not a durable model capability claim.

GPT 5.4 Mini — MotionBench