Bring Vision to Your Apps

Essential Vision Tools & APIs for Businesses & Developers

Access over 150 API endpoints → from one platform. PixLab brings together ID document scanning, agentic desktop automation, document intelligence, background removal, and a unified Vision Platform for developers and business teams. Build production workflows across cloud, browser, desktop, and on-premises deployments.

Open PixLab Console Start with PixLab
Quick Access
API Portal API Endpoints ID Scan & Extract Vision Platform Mobile Apps

Trusted by Product Teams and Developers Worldwide

Creative Zone - PixLab client
Media Hub - PixLab Client
Hub People - PixLab client
Symisc - PixLab Client
GVIC - PixLab client
Monthly Requests Served
1.5B+
Global API traffic. Monitor your usage in the PixLab Console ↗.
Uptime Reliability
99.9%
Infrastructure built for ongoing production workloads.
Active API Endpoints
150+
Find request formats and examples in the API catalog.
Unified API Platform

One Platform for Vision, Documents, and Media

Choose the APIs and tools your application needs: document extraction, image processing, content moderation and media editing. Add desktop automation with Think-Act or test document workflows in Vision Workspace, with documentation for each integration.

Open API Portal Browse Endpoints
Facial Authentication icon

FACEIO Facial Authentication

Enable passwordless sign-in and identity verification on web and mobile with FACEIO, powered by secure, privacy-first facial recognition.

Start with FACEIO
ID Scan & Extract API icon

DOCSCAN ID & Document Extraction

Parse passports, national IDs, and driver's licenses from 197+ countries into normalized JSON. Supports 11,000+ document types with lightweight integration.

Explore ID Scan API
Vision Workspace + VLM APIs icon

Vision Workspace + VLM APIs

Extract structured data, answer document questions, and automate image understanding with vision-language workflows powered by leading Vision Language Models.

Explore Vision Platform
Creative Tools icon

Creative and Media Tooling

Build faster with AI Photo Editor, AI Video Editor, Annotate, Creative Toolbox, Convert Box, and Bulk Image Background Removal in production-ready browser workflows.

Explore Creative Tools
ID Document Scanning API

DOCSCAN API for Global ID Document Scanning

Extract structured JSON from 11,000+ identity document types, including passports, driver's licenses, national IDs, and supporting documents across 197+ countries and territories. DOCSCAN is designed for document intake, onboarding forms, record processing, and back-office data entry.

  • Document Identification Instantly recognize and classify passports, national IDs, and driver's licenses before data extraction begins.
  • Precise Data Extraction Extract full name, ID number, date of birth, address, MRZ code, and more from any supported document in real time.
  • Privacy-First Processing All processing runs strictly in-memory. Zero persistent storage, zero retained images - enforced by default.
Read DOCSCAN Docs Create API Key
docscan.py
Python 
# ID Scan and Extract - Python code sample
# Endpoint: https://pixlab.io/endpoints/docscan

import requests

req = requests.get('https://api.pixlab.io/docscan', params={
  'img':  'https://example.com/passport.jpg',
  'type': 'passport',  # 'idcard' | 'passport' | 'driverslicense'
  'key':  'PIXLAB_API_KEY', # Get yours at console.pixlab.io
})

reply = req.json()

if reply['status'] != 200:
    print(reply['error'])
else:
    # Parsed fields returned from the document
    fields = reply['fields']
    print("Issuing Country: "  + fields.get('issuingCountry',  'N/A'))
    print("Full Name:        "  + fields.get('fullName',       'N/A'))
    print("Document Number:  "  + fields.get('documentNumber',  'N/A'))
    print("Date of Birth:    "  + fields.get('dateOfBirth',    'N/A'))
    print("Address:          "  + fields.get('address',        'N/A'))
    print("MRZ Code:         "  + fields.get('mrz',            'N/A'))
REST · No SDK required · JSON response Full reference →

MRZ + Non-MRZ Coverage

Process passports, ID cards, and licenses with or without MRZ using one consistent DOCSCAN workflow.

Normalized JSON Output

Return standardized fields such as full name, document number, dates, and address for direct use in backend and document intake workflows.

Face Crop + Quality Signals

Use extracted face crops and image quality indicators in downstream applications and manual document review workflows.

Image APIs & Local Models

Production Image APIs and Local Vision Models

Use focused REST endpoints for background and watermark removal, then explore small vision models in development for browser automation, OCR, and image understanding.

bgremove.py
Python 
# Background removal with a public image URL
import base64, requests

reply = requests.get(
    'https://api.pixlab.io/bgremove',
    params={
        'img': 'https://example.com/product.jpg',
        'key': 'PIXLAB_API_KEY'
    }
).json()

if reply['status'] != 200:
    print(reply.get('error', 'Request failed'))
else:
    img_data = reply['imgData']
    with open('bg-removed.png', 'wb') as f:
        f.write(base64.b64decode(img_data))
    print('Saved bg-removed.png')
INPUT
Input image for background removal API
OUTPUT
Background removed output example
GET /bgremove

Background Removal API

Isolate the main subject and remove the background with pixel-accurate segmentation. Ideal for product catalogs, ad creatives, profile photos, and automated media pipelines.

  • High-Precision Subject Segmentation Automatically detects foreground objects without manual masking, making output reliable across varied scenes.
  • Flexible Output Delivery Receive image data as base64 by default, with optional direct file link output when object storage integration is enabled.
  • Scalable Batch Workflows Pair with Bulk Image Background Removal for high-volume processing.
View BG-REMOVE Docs Open Bulk Tool
POST /txtremove

Text & Watermark Remove API

Remove text overlays, captions, and watermark artifacts using inpainting-based reconstruction. Output remains natural for downstream editing, publishing, or model ingestion workflows.

  • Automated Inpainting Reconstructs removed regions so cleaned assets can be reused in professional creative and content operations.
  • Simple POST Integration Upload local files as multipart form data and get clean output in the same JSON response pattern used by other PixLab endpoints.
  • Composable with Other APIs Chain cleaned output into OCR, translation, and document extraction flows for end-to-end automation.
View TXT-REMOVE Docs Create API Key
txtremove.py
Python 
# Remove text/watermark from a local image
import base64, requests

with open('./watermarked.jpg', 'rb') as image_file:
    req = requests.post(
        'https://api.pixlab.io/txtremove',
        files={'file': image_file},
        data={'key': 'PIXLAB_API_KEY'}
    )

reply = req.json()
if reply['status'] != 200:
    print(reply.get('error', 'Request failed'))
else:
    with open('cleaned.png', 'wb') as f:
        f.write(base64.b64decode(reply['imgData']))
    print('Saved cleaned.png')
WATERMARKED
Image with text watermark before cleanup
CLEANED
Image after text and watermark removal

PixWeb & PixVLM Small

PixWeb and PixVLM Small are compact vision models in development for browser, desktop and edge applications. PixWeb interprets browser screenshots for UI automation. PixVLM Small focuses on OCR, object recognition, visual questions, captions and image-to-JSON extraction.

Browser automation
Identify visible text, controls, forms and workflow blockers. Pass those observations to a planner or controller, then use a fresh screenshot to check the result of an action.
Image understanding
Extract text from receipts, labels and simple forms, recognize common objects, or request captions and named JSON fields for your application to validate.
Browser and edge deployment
WebGPU, CPU inference through WebAssembly, and compatible ONNX runtimes are deployment targets. Released profiles will document tested hardware, memory use and performance.
Explore PixWeb & PixVLM Small View on GitHub

PixWeb

Browser screenshots for UI automation.

Input
A browser-page screenshot
Intended output
  • Visible text
  • Controls
  • Forms
  • Page state
Application role
Visual observations for your planner and browser controller. Your application manages actions and permissions.

PixVLM Small

Image understanding for your application.

Input
An image
Intended output
  • Extracted text
  • Objects
  • Captions
  • JSON fields
Application role
Image extraction and question-answering workflows, with field validation handled by your application.

Tested releases will include model downloads, browser demos, compatibility details and measured benchmarks. Choose the model that fits your task; an application does not need to load both.

Desktop automation with Think-Act

Think-Act combines a desktop agent, writing assistance, OCR, document intelligence and screen recording in one Windows and macOS app. Use a tool directly or describe a multi-step task, provide files or screen context, and follow its progress with approval controls and reviewable activity.

Documents and OCR
Extract text and query PDFs, scans, receipts, screenshots and contracts for document review and data-entry tasks.
Writing and research
Draft, rewrite and summarize content, or use the agent for research with the tools available to the task.
Reusable workflows
Save instructions as Skills and reuse them for recurring document, writing and desktop jobs.
MCP integrations
Connect compatible tools and data sources through MCP when a workflow needs capabilities beyond the built-in tools.
Screen capture and desktop tools
Capture screenshots and recordings, work with files, and give the agent the context needed for a supported task.
Approval controls
Review required approvals, inspect activity logs and configure which supported actions are allowed, require confirmation or are blocked.
Explore Think-Act Download Think-Act
Think-Act
Think-Act Agentic Dashboard with task instructions, success criteria, Skills, MCP connectors and activity logs
Set task instructions, manage Skills and connectors, and review results and activity in the Think-Act workspace.

Media forensics and deepfake analysis

PixLab Forensic brings image forensics, deepfake investigation, recording review and text analysis into one workbench. Examine the original source, compare its provenance and diagnostic signals, and preserve the context behind your findings in projects and reports.

AI image and frame assessment
Assess an image or selected video frame with connected AI. Review the service-reported confidence, explanation and signals alongside your own observations.
Image diagnostics and provenance
Inspect EXIF/GPS, Content Credentials, Error Level Analysis, JPEG compression and noise-variance overlays to identify details that need closer review.
Audio and video review
Navigate waveforms and spectrograms, replay recordings and capture timestamped video frames while retaining their parent recording context.
Text and reference analysis
Search source passages, follow URLs, emails and dates, preserve text revisions and optionally request AI review of a limited excerpt.
Explore PixLab Forensic Open the Workbench
forensic.pixlab.io
PixLab Forensic Deepfake Detection workspace with AI assessment and image provenance tools
Image and selected-frame assessment in PixLab Forensic. A frame result applies to that frame, rather than the whole recording.
File comparison and relationships
Compare SHA-256 and MD5 hashes for exact matches, use dHash for visual similarity, and explore relationships built from examined sources and recorded metadata.
Saved investigations and reports
Retain sources, captured frames, notes and text revisions in a project. Resume the investigation and export a JSON evidence bundle or a print-ready report that can be saved as PDF.

Developer resources

From Your First API Request to a Working Integration

Choose an endpoint, try it with a sample file, and inspect the response before adding it to your application. Use the API references and practical tutorials for request parameters, response fields and error handling.

Generate Your First API Key

  1. Set up your API key

    Create your key in the PixLab Console and review the plan limits for your workload.

  2. Send a sample request

    Follow the REST API guide with a representative file. Check the returned fields, image or document.

  3. Build and test your workflow

    Use the tutorials and knowledge base for integration examples, batch processing and common implementation issues.

Use Cases & Workflows

Scalable & Unified APIs at Your Fingertips!

These examples reflect common production patterns across fintech, commerce, logistics, operations, and creator tooling. Use production-ready APIs, browser tools, and mobile apps to move from proof of concept to release with less integration overhead.

ID document scanning and structured field extraction using the DOCSCAN API
ID Scan API

ID Document Scanning

Identify passports, visas, and national IDs, then return normalized fields for onboarding forms and document intake workflows.

Open DOCSCAN Reference
Vision Platform document Q&A - asking questions over contracts and invoices
Vision LLM API

Document Q&A

Ask questions over contracts, invoices, forms, and reports - and return structured answers using vision-language models.

Explore Vision LLMs
Content moderation - NSFW detection and automatic blur applied to unsafe images
Content Moderation API

Content Moderation

Score image uploads for adult or graphic content using the NSFW endpoint, then automatically blur or reject unsafe assets.

Try NSFW Endpoint
Face detection and privacy blur applied to human faces in an image
Face Detection API

Detect & Blur Faces

Detect all human faces in an image or video frame, then apply a privacy blur - using just two PixLab API calls.

Try FACEDETECT
Background removal - subject isolated from image background using deep learning
Background Remove API

Product Image Background Removal

Remove backgrounds from product shots, portraits, and creative assets with a single API call - no manual masking needed.

Try BG-REMOVE
Convert Box web interface for media and document format conversion in the browser
Media Conversion

Media & Document Conversion

Convert between 239+ image, video, audio, and document formats in-browser or via API - no installs, no server setup, batch-ready.

Open Convert Box
Explore All API Use Cases
Authentication, Documents & Automation

Passwordless Auth, ID Document Scanning, and Desktop Workflow Automation

Combine FACEIO for passwordless sign-in, DOCSCAN for global identity document extraction, and Think-Act for supervised desktop and document workflows.

Authentication Framework

FACEIO: Passwordless Facial Authentication

Add facial authentication to an existing website or web app with FACEIO's JavaScript library. Enroll new users and authenticate returning users through their camera, while your backend keeps control of accounts and authorization.

Integrate FACEIO
FACEIO passwordless facial authentication flow for secure sign-in

Connect Your Business Workflows

Connect account sign-in, document intake and analysis to the systems your team already uses. Choose a separate tool for each part of the workflow.

  • Document Intake & Authentication Extract ID document fields with DOCSCAN, then add passwordless facial authentication with FACEIO ↗.
  • Invoice & Contract Review Ask questions over uploaded PDFs, contracts, and invoices using the Vision LLM API to find key clauses and figures for review.
  • Desktop Workflow Automation Use Think-Act for supervised work with documents, screenshots, research, writing, and repeatable desktop tasks.
Vision LLM API View All Endpoints

150+ API Endpoints for Production Workflows

Browse the PixLab API catalog for ID document scanning, image and video processing, OCR, content analysis, automation, and more. Each reference focuses on the request format, response fields, and integration details developers need.

  • 150+ documented endpoints in one searchable catalog
  • Request parameters, response fields, error handling, and code samples
  • APIs for documents, images, video, moderation, OCR, and automation
Browse API Endpoints Open PixLab Console
PixLab API endpoint catalog and developer documentation
Vision Platform

Test Document Workflows in PixLab Vision Platform

Try document questions, OCR and structured extraction with your own files. Review the results in Vision Workspace, then use the API reference to connect the workflow to your application.

PixLab Vision Platform workspace - testing OCR and VLM workflows with real document payloads

From Document Preview to API Request

Compare model responses, check extracted fields and work through representative documents before writing your integration.

  • Upload real files and test OCR, VLM parsing, and document Q&A in one workspace.
  • Validate extracted fields and structured outputs before writing production integration code.
  • Move directly from workspace testing to endpoint docs, request examples, and implementation references.
Open Vision Workspace
Enterprise

On-Premises and Enterprise Deployment

Explore deployment on your own infrastructure, including private cloud and air-gapped environments. Review supported APIs, capacity and service requirements with the PixLab team.

  • Air-gapped & private cloud deployment
  • Covers ID Scan, OCR, Vision LLM & Media APIs
  • Deployment and support options for your requirements
Explore On-Premises
Mobile Apps Hub

Scan, Edit and Translate with PixLab Apps for iOS

Explore PixLab iOS apps for video editing, photo editing, document scanning and camera translation in the Mobile Hub. Find features, screenshots and download links for each app.

PixLab Photo Editor

Prompt-based photo editing on iOS, with background removal, filters, layers and creative tools for everyday image work.

View iOS Product Page
AI Photo Editor - iOS app interface showing creative editing tools

World Lens

Point and translate - live on-screen translation of text in the real world, powered by PixLab OCR and translation APIs.

View iOS Product Page

Find Your Next Mobile Tool

Browse the mobile apps for editing, scanning and translation. Developers can explore the API catalog to add related vision and media features to their own products.

Open Mobile Apps Hub Build with PixLab APIs
Open Source SDKs and Native Libraries

Build Locally with Native C and C++ Libraries

Run vision, inference, and data workflows in embedded, desktop, or edge environments when local execution and runtime control matter.

SOD

Embedded computer vision and machine learning library in C for real-time object detection and face recognition in CPU-first environments.

Learn more GitHub

SyNumPy

C++ NumPy file reader/writer - read and write .npy and .npz array files from C++ without Python, for ML pipelines.

Learn more GitHub

Tiny Dream

Stable Diffusion inference in pure C/C++ - generate images from text prompts on CPU without Python or PyTorch.

Learn more GitHub

ART

Real-Time ASCII Art C Library - transform images and video frames into ASCII art at runtime with configurable density.

Learn more GitHub

Explore PixLab's open source libraries and start building locally.

Open Downloads

Find your starting point

Choose the Tools for Your Project

Start with the job you need to do. These links take you to the product details, API documentation and deployment options for each workflow.

Extract and Query Document Data

Test document questions, OCR and structured extraction in the Vision Platform, then integrate the VLM API with your application.

Add Face Login or Scan IDs

Add passwordless facial login with FACEIO, and extract structured fields from 11,000+ document types across 197+ countries with DOCSCAN.

Automate Desktop Tasks

Use Think-Act for supervised document work, screenshots, research, writing, and repeatable tasks across Windows and macOS.

Plan a Private Deployment

Explore on-premises deployment for your infrastructure. Discuss supported APIs, air-gapped environments, dedicated capacity and service requirements with the team.

FAQ

Common Questions About PixLab APIs & Tools

Straightforward answers for teams evaluating PixLab or planning an integration.

Browse API Endpoints
What can businesses and developers build with PixLab?
PixLab provides APIs and tools for ID document scanning, image and video processing, media forensics, document analysis and desktop automation. Developers can integrate APIs, while business teams, marketers and creators can use the browser, desktop and mobile tools.
How do I start using the PixLab APIs?
Create an API key in the PixLab Console, choose an endpoint from the API catalog, and follow the REST API guide to send your first request. A 7-day free trial lets you evaluate the APIs before choosing a monthly plan.
Does DOCSCAN verify identity or authenticate ID documents?
No. DOCSCAN identifies supported document types and extracts structured fields from passports, ID cards, driver's licenses, visas, and related documents. It does not authenticate documents, verify identity, or perform KYC or AML checks.
Can PixLab support high-volume image and media processing?
Yes. The background removal API and text and watermark removal API can be integrated into bulk image workflows for catalogs, stores and marketing assets. Check monthly plans for usage limits and confirm throughput requirements for your chosen endpoint.
What are PixLab Forensic, PixWeb, and PixVLM Small?
PixLab Forensic brings together AI image assessment, image forensics, video frame review, audio spectrograms, text analysis, and investigation reports. PixWeb and PixVLM Small are in development for browser automation, OCR, image understanding, and local inference.
Where can I find tutorials and code examples?
Visit the Resources library for tutorials and knowledge base articles on AI, computer vision, APIs and developer tools. Current guides cover DOCSCAN, background removal, text and watermark removal, Think-Act, FACEIO and video editing.
Can I deploy PixLab on my own infrastructure?
On-premises deployment is available for teams that need to run supported PixLab APIs on their own infrastructure. Review the supported services, hardware requirements, deployment options and support arrangements with the PixLab team.

Build with PixLab

Start with One API Call

Open the PixLab Console to get your API key and start a 7-day free trial. Choose an endpoint, try it with your files, and review the output before building it into your workflow.

Start Building with PixLab View Pricing Plans

Monthly plans are available. Usage and throughput limits depend on your plan and endpoint.