What is Horay.ai?

Horay.ai delivers an efficient and user-friendly platform specializing in high-speed inference for a variety of Artificial Intelligence models. It offers access to scalable Large Language Models (LLMs) such as Llama3, Mixtral, Qwen, and Deepseek, featuring out-of-the-box acceleration capabilities. The platform also includes a diverse range of Embedding and Reranker models designed to enhance Retrieval-Augmented Generation (RAG) processes, making them more efficient and straightforward.

Designed for developers, Horay.ai enables seamless integration of its fast model services with just a single line of code. It supports text-to-image and text-to-video model integration, including SDXL and photomaker, along with accelerated Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) models for low-latency voice generation. The platform's ultra-low latency API facilitates the development of responsive applications like interactive Agents and Chat2DB tools, while optimized APIs contribute to reduced costs for image generation tasks.

Features

High-Speed Inference: Delivers efficient, user-friendly, and scalable model inference with acceleration.
LLM Access: Provides access to models like Llama3, Mixtral, Qwen, Deepseek.
Embedding/Reranker Models: Offers models to improve RAG efficiency.
Image & Video Generation Models: Includes text-to-image/video models like SDXL, SDXL lightning, photomaker, instantid.
Voice Generation Models: Offers accelerated ASR/TTS models for low-latency voice synthesis.
Simple API Integration: Allows integration with a single line of code.
Low Latency API: Enables development of fast-interacting applications (Agents, Chat2DB).

Use Cases

Developing AI-powered applications requiring fast LLM responses.
Building Retrieval-Augmented Generation (RAG) systems.
Integrating text-to-image generation capabilities into apps.
Creating text-to-video features.
Implementing real-time voice generation or speech recognition.
Developing interactive AI agents or chatbots.
Building database interaction tools using natural language (Chat2DB).
Optimizing costs for AI model usage in applications.

FAQs

How much does Horay.ai cost?

Horay.ai uses a pay-as-you-go model based on token usage for serverless inference or per GPU usage time for on-demand deployments. New users receive free credits. For Enterprise pricing, please contact support@horay.ai.
Do you offer SLAs for Serverless usage?

No, the multi-tenant serverless offering does not come with Service Level Agreements (SLAs).
How do I get started with Horay.ai?

Sign up for an account at https://dash.horay.ai to receive free credits and start using serverless inference or on-demand deployments. Contact support@horay.ai for Enterprise-grade requirements.
Are there discounts for bulk spend on serverless deployments?

Horay.ai's publicly accessible services have standard rates for all customers; specific bulk discounts for serverless usage are not mentioned.
What are rate limits and how can they be increased?

Rate limits restrict API access frequency based on metrics like requests per minute (RPM), requests per day (RPD), tokens per minute (TPM), tokens per day (TPD), and images per minute (IPM). Exceeding limits results in rejected requests. Rate limits are automatically increased based on historic spend tiers.