Mighty Inference Server

Instantly launch a powerful NLP web service in your own stack.

Text in, Inference out

Embeddings, Question Answering, Classification, and more for any product.

Read more

Really really fast

Unlimited requests at less than half your current latency, for one low predictable price.

Read more

Ready for Production

The ready-to-use inference solution you've been looking for. Plug it into your stack, hassle free.

Read more

Mighty is a perfect AI microservice.

Deploying AI to production is expensive and painful. But it shouldn't be. We believe integrating AI should be predictable, easy, and freely scalable.

We built Mighty Inference Server to take the pain out of model hosting and performance tuning, to automatically give you the highest possible throughput, to empower you to scale with predictable costs, on your terms in your stack.


Download here!

Mighty Ecommerce Embeddings

Introducing Mighty Ecommerce - the best embedding model for your store's site-search.

Enables customers to find the products they're looking for more quickly, leading to higher satisfaction rates and increased sales.

High Precision and Recall. Outperforms other embedding models by a wide margin.

Really Fast, no GPU needed! Up to 180 embeddings per second on a single CPU core.

Flexible experience.Designed for Mobile, Tablet, and Desktop result pages.

Deploy in your stack with zero lock-in. Build your store on your terms, not someone else's expensive API.

Easily plug it in to your existing site-search with Mighty connectors!


Connects to any Vector Search Engine!

Case Study: Vectorizing the Code of Federal Regulations

  • 2,071,617 total paragraphs
  • 522MB of text
  • 1538 paragraphs per second
  • 19GB vectors generated in 22 minutes
  • Cloud costs: $3.00

In this study we show how to quickly get sentence-transformer embeddings for the entire eCFR (electronic Code of Federal Regulations).

The reading and lab takes less than 40 minutes to generate all vectors in an AWS volume, that can be used further for semantic search.

Quick Start

Step 1. Install
docker pull maxdotio/mighty
docker run -p 5050:5050 -t maxdotio/mighty
>Mighty server for embeddings is listening on http://localhost:5050
Step 2. Infer!
curl http://localhost:5050/\?text=Hello+mighty.
>{
>  "took":8,
>  "text":"Hello mighty.",
>  "shape":[5,384],
>  "outputs":[[0.066911593079,0.162673592567,...,-0.233405888080]]
>}
Step 1. Download
curl http://max.io/mighty-mac.tar.gz --output mighty-mac.tar.gz
tar -zxf mighty-mac.tar.gz
Step 2. Run
mighty/mighty
>Mighty server for embeddings is listening on http://localhost:5050
Step 3. Infer!
curl http://localhost:5050/\?text=Hello+mighty.
>{
>  "took":8,
>  "text":"Hello mighty.",
>  "shape":[5,384],
>  "outputs":[[0.066911593079,0.162673592567,...,-0.233405888080]]
>}
Step 1. Download
sudo apt-get install curl hwloc libhwloc-dev
curl http://max.io/mighty-linux.tar.gz --output mighty-linux.tar.gz
tar -zxf mighty-linux.tar.gz
Step 2. Run
mighty/mighty
>Mighty server for embeddings is listening on http://localhost:5050
Step 3. Infer!
curl http://localhost:5050/\?text=Hello+mighty.
>{
>  "took":8,
>  "text":"Hello mighty.",
>  "shape":[5,384],
>  "outputs":[[0.066911593079,0.162673592567,...,-0.233405888080]]
>}
Step 1. Download
curl http://max.io/mighty-windows.zip
unzip mighty-windows.zip
Step 2. Run
mighty\mighty.ps1
>Mighty server for embeddings is listening on http://localhost:5050
Step 3. Infer!
curl http://localhost:5050/\?text=Hello+mighty.
>{
>  "took":8,
>  "text":"Hello mighty.",
>  "shape":[5,384],
>  "outputs":[[0.066911593079,0.162673592567,...,-0.233405888080]]
>}

Decision Making Table

Mighty Build your own Paid APIs
Low-Latency - -
Production Ready -
Unlimited Requests -
Unlimited Scalability -
Cost $99 per month Months of developer time Pay-per-use (expensive!)

Pricing

Popular

Free

FREE

  • Unlimited non-commercial use
Download for free

Commercial

$99/mo

  • Unlimited deployments for business
Order Now
New

Cloud Providers

Let's talk

  • Partner with MAX.IO
Order Now