Hive AI

Overview

Hive is a cloud-based AI platform that provides businesses with advanced tools to understand, search, and generate content across text, image, video, and audio formats. By offering pre-trained models accessible through simple API integrations, Hive enables organizations to enhance content moderation, improve user experiences, and streamline workflows.

Key Features

Content Moderation: Detect and filter harmful content, including NSFW material, hate speech, and violence, across various media types.
Brand Protection: Monitor and protect intellectual property by identifying unauthorised use of logos, trademarks, and likenesses.
Fraud Detection: Uses AI to identify AI-generated content, phishing attempts, and counterfeit media to safeguard against fraud.
Generative AI: Create and transform content with cutting-edge generative models for text, image, and video generation.
Compliance Monitoring: Ensure adherence to safety standards and regulations by detecting policy violations and implementing age verification.

Pros

Scalability: Handle large volumes of content efficiently with Hive's robust AI models.
Ease of Integration: Seamlessly incorporate Hive's APIs into existing systems with minimal setup.
Comprehensive Coverage: Address a wide range of content-related challenges, from moderation to generation.
Real-Time Processing: Benefit from fast and accurate content analysis and generation.

Use Cases

Social Media Platforms: Implement content moderation to maintain community standards and safety.
E-Commerce Websites: Enhance product listings with AI-generated descriptions and images.
News Organisations: Use AI for content tagging, summarisation, and moderation.
Educational Institutions: Deploy AI tools for content analysis and accessibility improvements.

Frequently Asked Questions

What's the difference between the V2 and V3 APIs?

V3 API: Self-serve, "instant-on" API with default rate limits (e.g., 100 requests/day for Text and Visual Moderation). Ideal for quick integration.
V2 API: Designed for Enterprise projects requiring higher throughput, custom support, or invoiced billing. A Hive Sales representative must enable a V2 project before you can obtain a V2 key.

Which media types can Hive classify/moderate?

Hive's models support text, images, videos, audio, and text within images (OCR).

Should we submit our tasks to the API synchronously or asynchronously?

Synchronous submission: Optimised for real-time moderation needs. If you have a continuous flow of tasks with low latency requirements and need responses quickly, use synchronous submission.
Asynchronous submission: Preferred for large volumes of tasks, tasks referencing large files (e.g., long videos or audio files), or projects using Hive manual review.

Does Hive support batch processing?

Currently, Hive APIs do not support batch processing. However, our APIs achieve high concurrency in task processing with generous default rate limits of 25-50 tasks per second, depending on the model. At full use, this yields 2-4 million completed tasks per day.

What happens to our data after tasks are processed by Hive APIs?

Our default policy is to retain customer data for 14 days. We can increase or decrease our retention time upon request according to your needs (including not retaining your data at all). Please note that increasing retention time may incur additional charges.

Can we access multiple models with a single task submission/API call?

Currently, we can combine visual moderation and contextual scene classification into a single API endpoint, enabling customers to access both model outputs with a single task. In other cases, classification across multiple formats requires separate tasks to be submitted to the API for each model.