Photoroom has published results from a randomised test with Vestiaire Collective in which listings with their main photo edited by Photoroom sold at a 5% higher rate of liquidity than listings edited by the resale platform's in-house tool.
The test was randomised. New listings were split into two groups, one processed by Photoroom and one by the clipping tool Vestiaire Collective built years ago. Product pages in the Photoroom group drew 3 to 7% more unique views, with bags and shoes gaining most. Performance marketing ads using the same photos became 2% more efficient, and digital authentication was unaffected. It took eight weeks from deciding to test to having results.
Fidelity was the deciding factor. Vestiaire Collective authenticates some items from photos alone, checking labels, stitching and the way light falls on the material, so a model that smooths a scuff or redraws a logo breaks the process. Thomas Hezard, Chief Product Officer at Vestiaire Collective, said: “We tested different GenAI solutions, and each time they changed some details. Then it was no longer possible to authenticate from the pictures or to make sure buyers could trust them. Photoroom maintained product fidelity, and that's clearly the reason we chose it.”
Photoroom's own benchmark of 4,250 AI-generated images across 850 products found the best image-editing models passed a full fidelity check 29% of the time. It is the vendor's benchmark, not an independent one.
In production the Photoroom API removes the background from the main photo, adds a shadow, centres the item and corrects the lighting before each listing goes live. Vestiaire Collective operates in 70 countries and adds around 40,000 listings a day. The pipeline combines Photoroom's models with open models including FLUX.2 [klein] and DINOv2, runs on NVIDIA H100 and H200 GPUs with capacity from Nebius, and uses TensorRT and Triton. Using NVIDIA Dynamo Snapshot, Photoroom cut the time for a new GPU server to come online from about three and a half minutes to about one.
To stay across the latest in cloud, AI and enterprise tech analysis from Compare the Cloud, subscribe to our weekly newsletter at https://www.comparethecloud.net/newsletter
