Meteron AI

Core Features & Characteristics

• Metering: Provides a simple yet powerful metering mechanism. Charge users per request or per token (or both). Supports per-user metering by specifying daily and monthly limits when adding model endpoints; adding the X-User header with user ID/email ensures the user cannot exceed limits.
• Elastic scaling: Queue up and load-balance requests across servers. Add more servers at any time to absorb high demand spikes. Supports server concurrency control, intelligent QoS, performance tracking, automatic load balancing (automatic retries coming soon).
• Unlimited Storage: Uploads images to the cloud so users will not run out of storage. All major cloud providers are supported. Custom Cloud Storage (your own S3, GCS, Azure Storage, etc.) coming soon.
• Any model - text, image: Works with any model. Plug in Llama, Mistral, Stable Diffusion v1, v2, DALL-E, or any other image generation model.
• Queue prioritization: Default standard business rules. Specify priority class with each request: high (VIP users, no queueing delays), medium (incurs delays but ahead of low), low (served last, for "free" users when there's no load).
• Credit system: Supports a credit system for billing and limits.

Target Users & Core Advantages

• Target Users: Developers and teams looking to build and monetize AI-powered products.
• Core Advantages: An all-in-one AI toolset that frees developers from time-consuming, unnecessary processes, allowing them to focus on building AI products. Build AI apps in hours, not weeks. It is a "low-code" service requiring only HTTP knowledge, supporting favorite HTTP clients like curl, Python requests, and JavaScript fetch.

Typical Use Cases

1. Lightweight AI app: An image gallery generator powered by Stable Diffusion XL, built with Vuetify, Lightning AI, and Meteron.
2. Multi-tenant application: Users register and generate room images using the Controlnet AI model, with assets, queueing, and metering enforced by Meteron.
3. Python integration: A collection of individual functions for sending image generation requests, querying results, ensuring per-user limits, and charging per request/tokens.

Ecosystem of Integrations

Supports integrations with PyTorch Lightning, TensorFlow, Python, OpenAI, AWS, etc.

Pricing Information

Freemium model with monthly and yearly (10% off coming soon) billing:

• Free ($0/mo): 1 Admin, 5GB File Storage, 1,500 Image Generations, 10,000 LLM chat completions.
• Professional ($39/mo): 5 Admins, 300GB File Storage, 10,000 Image Generations, 50,000 LLM chat completions.
• Business ($199/mo): 30 Admins, 2TB File Storage, 100,000 Image Generations, 800,000 LLM chat completions. Additional storage and generations can be purchased. Accepts all major credit cards and direct wire transfers. On-prem licenses are also available.

FAQs

• How to tell Meteron where servers are: Through the web UI for static servers or via a simple API to update servers dynamically in real time (for platforms like Lightning AI, RunPod).
• Self-hosting: Yes, on-prem licenses are available to run on any cloud provider.
• Payment methods: All major credit cards and direct wire transfers.