v3

latestOpenAPI 3.1.0raw.githubusercontent.com2026-07-2131914.7 KB

Vision

Run Locate

post/api/v1/vision/locate

Request body

imagestring required

Base64 encoded image (PNG or JPEG)

modelstring nullable

VLM model to use; must be one of the models from GET /inference/models. Omit to use the server's configured default. The system prompt is fixed to the element-locator task.

querystring required

Natural-language target description

Response

Successful Response

completion_tokensinteger

Completion (output) tokens the model billed for this call

confidencenumber required

Model self-reported confidence

cost_microdollarsinteger required

Provider-reported cost in microdollars

foundboolean required

Whether the VLM located the target

latency_msinteger required

End-to-end VLM call latency

matched_text_indexinteger nullable

Index into the request texts list when the answer is OCR-grounded.

modelstring required

Model identifier returned by the provider

prompt_tokensinteger

Prompt (input) tokens the model billed for this call