Skip to content
runcrate
Console

FACE DETECTION API

Detect faces with a single API call.

Send a photo and a question, get faces, bounding boxes, age estimates and expressions back as JSON, from vision-language models on the chat API.

Get an API keySee model prices

Hand this to your agent

Add a Runcrate model to this project: run curl -fsSL"https://runcrate.ai/agents.md?intent=build"and follow it.

Your agent asks for an API key. Create a key

Works inClaude CodeCodexOpenCodeClineRoo CodeKilo CodeCursor
InputImage + question
FormatVision + chat
OutputStructured JSON

Quick start

Hand it to your agent. Or read the request it sends.

Your agent writes and runs it. The same key works for every model and every GPU.

Docs
Count the faces in every photo in ./event with Qwen3 VL on Runcrate and write a CSV of file and count. Runs curl https://api.runcrate.ai/v1/chat/completions \ -H "Authorization: Bearer rc_live_YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "Qwen/Qwen3-VL-235B-A22B-Instruct", "messages": [{ "role": "user", "content": [ {"type": "image_url", "image_url": {"url": "https://example.com/photo.jpg"}}, {"type": "text", "text": "Return face bounding boxes, ages and expressions."} ] }] }'.

Available models

Models you can call today. Billed prices, as charged.

ModelMakerPriceDetail
Qwen/Qwen3-VL-235B-A22B-InstructAlibaba$0.21 in, $0.924 out per 1M tokens235B MoE, detailed visual understanding
Qwen/Qwen3-VL-30B-A3B-InstructAlibaba$0.158 in, $0.63 out per 1M tokens30B MoE, efficient processing
See every model price

Why Runcrate

Built for production. One key, one balance.

  • Vision-language detection

    Use the full power of vision-language models for face detection. Ask for bounding boxes, age estimates, expressions, or any custom attribute in natural language.

  • Structured output

    Request JSON-formatted results with bounding boxes, confidence scores, and attributes. Parse programmatically for integration into your pipeline.

  • No specialized library

    Skip OpenCV, dlib, and MediaPipe setup. Use the same chat completions API you already use for text, with an image added.

  • Custom analysis

    Beyond detection: ask the model to analyze facial expressions, estimate demographics, count people, or describe the scene context around each face.

FAQ

Common questions.