Recorded model output
Damage detection
Visible damage, located on the source image.
/v1/damage/detect →Layer 2 · Models
Each endpoint is priced and rate-limited on its own. Call one in isolation, or chain the set — nothing assumes you adopted the others.
23 POST · one model each · async by default · webhook or poll
Recorded model output
Visible damage, located on the source image.
/v1/damage/detect →The contract
Inference is not instant, so nothing here pretends to be synchronous. All 23 return a job, and one status call reads every one of them — queued, computing, done, failed or timeout. Poll it, or take the webhook.
It is platform rather than inference — it runs no model, costs nothing and is not counted among the 23.
the status call
free
GET /v1/jobs/{id}
inference orchestration · no model
queued · computing · done · failed · timeout — plus webhook push
The reference
Priced and rate-limited one by one. The meter on the right is what a single call costs you — per vehicle, per image, per video or per detection.
Identity, attributes and a clean capture — everything downstream reads from here.
/v1/vehicle/poseper imageTools allowing the detection of the car as well as its position.
vehicle-pose/v1/vehicle/segmentper imageTool to perfectly segment the car.
car-segmentation/v1/vehicle/identifyper imageTool to identify the make, the model and the version of a vehicle from a photo.
model-detection/v1/vin/readper imageTool for reading the VIN of a vehicle.
vin-reader/v1/plate/readper imageTool for reading license plates.
plate-reader/v1/document/readper imageTool for reading vehicle documents.
document-reader/v1/dashboard/readper imageTool for reading the mileage and the warning lights from a dashboard photo.
dashboard-reader/v1/engine/detectper vehicleTool to detect internal engine damage from the sound of the engine.
engine-detectionFind it, then grade it, so your own rules can act on the result.
/v1/damage/detectper vehicleTool for detecting damage to the interior and exterior of the vehicle.
damage-detection/v1/damage/severityper detectionTool to assign a severity to a damage.
severity-estimation/v1/tire/readper imageTool for reading the type of a tire from its sidewall.
tire-reading/v1/tire/detectper imageTool for detecting tire wear.
tire-detectionTurns “there is damage” into “there is damage on this component” — what makes every downstream estimate defensible.
/v1/parts/recognizeper imageTool to give a name to each part of your vehicle.
parts-recognition/v1/parts/segmentper imageTool to segment a vehicle with its multiple parts.
parts-segmentation/v1/parts/decomposeper vehicleTool for decomposing a part of the vehicle into components.
parts-decompositionValuation and repair cost, computed from what was actually observed.
/v1/pricing/vehicleper vehicleTool to estimate the price of a vehicle in real time according to many criteria.
car-pricing/v1/pricing/repairper vehicleTool allowing to give the price of repairing a damage.
reparation-pricingExposed as raw infrastructure, not a batch retouching service. Call it per asset, in your own pipeline.
/v1/render/backgroundper imageTool to replace the background of images containing vehicles.
image-backgrounding/v1/render/enhanceper imageTool to improve the quality of an image.
quality-improver/v1/render/plateper imageTool for covering a license plate with a logo.
plate-replacer/v1/render/360per vehicleTool to generate a studio 360 of a vehicle from four photos.
car-360-generation/v1/render/videoper videoTool to generate a custom advertising video of a vehicle from photos.
ad-video-generation/v1/render/hotspotsper vehicleTool to anchor points of interest onto named parts of a vehicle across a set of images.
viewer-hotspoterThe rest of the stack
Send a photo to /v1/vehicle/pose and read the response. The other 22 are the same key.