We built BlueVetaUpright because general vision models are specifically bad at telling 90° from 270° on product photos (confirmed independently in RotBench, EACL 2026, and in our own 400-photo benchmark against ChatGPT: 91.1% vs 75.9%). Wanted to show it working on an actual photo instead of just citing numbers, so here's a real, unedited run: a real coffee-machine photo, deliberately rotated 90°, through the free /upright demo.
That's the real API response, not a mockup — rotationClockwise4Axis: 270, confidence score, latency, all real. Free to try with your own photo, no signup: blueveta.com/upright. Happy to answer questions about how the 4-axis + fine-tilt arbitration works.
Top comments (0)