Intelligence

Make your video, image, and audio files more useful

Extract rich insights from your media files

Search box and grid of image resultsSearch box and grid of image results

Best-in-class AI

We continuously review models so you always get state-of-the-art image and video intelligence without having to chop-and-change providers.

Powerful automation

Set-up powerful intelligence workflows using a simple declarative API. No more queues, retries, or monitoring to worry about.

Scale up effortlessly

Ittybit's infrastructure team have built broadcast-scale systems so you can build without thinking about GPUs, kubernetes, or serverless timeouts.

Don't go broke!

Predictable and affordable pricing means you can scale your intelligence workflows without burning through your entire seed round.

POST /tasks

Add media intelligence to your app in minutes

Description
View Docs

Make content searchable, improve your recommendations, and generate helpful alt-text with a description powered by computer vision.

Image by Alex Knight: https://unsplash.com/photos/vaA6EQiUSo4
const task = {
  url: "https://your.domain/image-to-analyze.jpg",
  kind: "description",
}

const response = await fetch('https://api.ittybit.com/tasks', {
  method: 'POST',
  headers: {'Authorization': `Bearer ${ITTYBIT_KEY}`},
  body: JSON.stringify(task)
});
{
  "kind": "description",
  "text": "A vibrant nighttime street scene in a bustling entertainment district of Japan. The street is awash in neon lights and colorful signage, creating a dazzling, cyberpunk-like atmosphere. Large illuminated signs advertise various businesses, including Bariki, a Thai yoga massage parlor, and Hyakkendana. The buildings are densely packed, with multiple floors of establishments visible. The street level is populated with a few pedestrians, their silhouettes visible against the bright lights. The color palette is dominated by vivid pinks, blues, and teals, giving the scene a surreal, electric feel. The overall impression is one of a lively, perhaps slightly chaotic, nightlife area full of energy and activity."
}
Speech
View Docs

Automatically transcribe your audio and video files to make them searchable and to support users watching on mute or with hearing impairments.

Image by Sebastian Pandelache: https://unsplash.com/photos/dZEEhW6L6E4
const task = {
  url: "https://your.domain/video-to-analyze.mp4",
  kind: "speech",
  speakers: true,
}
{
  "kind": "speech",
  "detected": true,
  "timeline": [
    {
      "start": 0,
      "end": 10,
      "text": "Hello, and welcome to UkeTube. I'm John Doe.",
      "speaker": 0,
      "confidence": 0.98
    },
    {
      "start": 10,
      "end": 20,
      "text": "And I'm Jessie Doe. Today we're going to learn how to play Viva La Vida by Coldplay.",
      "speaker": 1,
      "confidence": 0.95
    }
  ]
}
Labels
View Docs

Detect and label objects, people, and brands in your images and videos. Great for building custom search and recommendation systems or for discovering new marketing partners.

Image by Tobias Rehbein: https://unsplash.com/photos/FhnX6PDy9bA
const task = {
  url: "https://your.domain/video-to-analyze.mp4",
  kind: "objects",
}
{
  "kind": "objects",
  "detected": true,
  "items": [
    {
      "label": "Nike soccer cleats",
      "confidence": 0.85,
      "box": {"x": 300, "y": 450, "w": 200, "h": 100}
    },
    {
      "label": "Uhlsport soccer ball",
      "confidence": 0.95,
      "box": {"x": 450, "y": 300, "w": 100, "h": 100}
    },
    {
      "label": "White soccer shorts",
      "confidence": 0.97,
      "box": {"x": 100, "y": 100, "w": 300, "h": 200}
    },
    {
      "label": "Red advertising board",
      "confidence": 0.97,
      "box": {"x": 0, "y": 0, "w": 800, "h": 100}
    },
    {
      "label": "White soccer socks",
      "confidence": 0.98,
      "box": {"x": 200, "y": 300, "w": 150, "h": 250}
    },
    {
      "label": "Soccer field (grass)",
      "confidence": 0.99,
      "box": {"x": 0, "y": 400, "w": 800, "h": 200}
    },
  ]
}
Summarize
View Docs

Save hours of manual work by turning your uploads into articles with one API call. Get title, tag, and summary text suggestions ready to go straight into your CMS.

Image by Teemu Paananen: https://unsplash.com/photos/bzdhc5b3Bxs
const task = {
  url: "https://your.domain/audio-to-analyze.mp3",
  kind: "summary",
}
{
  "kind": "summary",
  "title": "Designing at Scale: Airbnb's Journey with Design Systems",
  "tags": ["Design systems", "Airbnb", "Product development", "Scalability", "UI/UX", "Accessibility", "React Sketchapp", "Organizational change"],
  "text": "Karri Saarinen, formerly of Airbnb, presents a compelling case for design systems as a solution to the increasing complexity of product development in the digital age. /\n/\nHe argues that design systems are not just about aesthetics, but are crucial tools for organizational efficiency, product consistency, and scalability./\n/\nSaarinen delves into Airbnb's approach to design systems, emphasizing the importance of functional components and primitives. He demonstrates how their system allows designers and engineers to work more efficiently across multiple platforms, resulting in a more cohesive user experience. The talk also touches on the process of maintaining and evolving the design system, highlighting the balance between core components and team-specific experiments./\n/\n### Beyond Aesthetics: The Organizational Impact/\n/\nWhat sets this talk apart is Saarinen's focus on the organizational aspects of implementing a design system. He stresses that success depends not just on the technical implementation, but on how well the system aligns with a company's goals and structure. This perspective elevates the conversation beyond mere design tools to a discussion of how design systems can drive broader organizational change and product strategy./\n/\n### Innovations and Future Directions/\n/\nThe presentation concludes with exciting innovations spawned by Airbnb's design system, including React Sketchapp. This open-source tool allows rendering of React components directly in Sketch, blurring the lines between design and development. Saarinen hints at the potential for such technologies to revolutionize the design process, enabling real-time language translations and responsive design previews directly within design tools."
}

Automatically detect adult content from your uploads. Essential for user-generated content and anybody building inclusive (and legally compliant) platforms.

Image by Luiz Rogerio Nunes: https://unsplash.com/photos/jibqNZ61Y98
const task = {
  url: "https://your.domain/image-to-analyze.jpg",
  kind: "nsfw",
  threshold: 0.6,
}
{
  "kind": "nsfw",
  "detected": true,
  "items": [
    {
      "category": "nudity",
      "confidence": 0.82
    }
  ]
}

Get started for free

10GB tasks. 100GB storage. 1TB delivery

Ittybit lets us handle user-generated content without worrying about weird formats, malicious files, or inappropriate material

Becky Fatemi

Becky Fatemi

Founder, BlackBook