Cerebrium provides prebuilt models that you can deploy to an API seamlessly

Deploying Prebuilt Models

cerebrium

Dashboard

Blog

Community

Status

Pricing

Cerebrium

Sign Up

Contact Us

Cerebrium is an AWS Sagemaker alternative providing all the features you need to quickly build an ML product.

Introduction

Cerebrium is our Python framework that meets you where you work - whether it be in Notebooks or .py files we want you to feel comfortable and be environment agnostic.

Installation

Cerebrium follows security best practices

Security

Reliability and Scalability

A list of hardware that is available on Cerebrium's platform.

Available Hardware

How to calculate the cost of your deployment on Cerebrium

Calculating compute cost

Deploy any Python code to serverless CPUs/GPUs giving you fine-grain control

Track the process of your model's build for a smooth deployment experience

Build Status

Know your model's current state for efficient deployment and management

Model Status

Using the CLI to quickly initialise a Cortex project.

Init Cortex Project

Persistent Storage for Caching

Using Secrets

Async Functionality

Using config files to configure Cortex deployments easily and quickly

Cortex Config Files

Long Running Tasks

Control the way your deployment scales up and down

Scaling of Deployments

To deploy your own custom version of Stable Diffusion

Stable Diffusion Example

To deploy Meta's segment anything model to classify clothes

Segment Anything Example

To deploy a Q&A application around a YouTube video

Langchain Q&A YouTube

To deploy your machine learning model to Cerebrium takes just 2 lines of code.

Deploying Model Ensembles

Persistent Memory in Conduit Processing Functions

Pre & Post Processing Functions

Saving File Output

Test locally

Sending File Input

Using Webhook Endpoints for Async Results

This example provides the steps to deploy a simple Onnx model from scratch using Cerebrium.

Onnx

This example provides the steps to deploy a simple Pytorch model (CNN) from scratch using Cerebrium.

Pytorch

This example provides the steps to deploy a simple Sci-Kit Learn model from scratch using Cerebrium.

Scikit

This example provides the steps to deploy a simple Spacy model (from scratch!) using Cerebrium.

Spacy

This example provides the steps to deploy a simple Tensorflow model from scratch using Cerebrium.

Tensorflow

This example provides the steps to deploy a simple XGBoost model (from scratch!) using Cerebrium.

XGBoost

Quickly and conveniently fine-tune your LLMs on with just one line of code!

Training

Advice on how best to structure your datasets

Dataset Curation

Guide for the config files for fine-tuning a large language model on Cerebrium

Config Files

Guide to prompt templating when fine-tuning your language model on Cerebrium.

Prompt Templating

Quickly and conveniently fine-tune your diffusion model with just one command!

Guide to fine-tuning a diffusion model on Cerebrium

Training Diffusion Models

Guide to curating a dataset for success when fine-tuning your diffusion model on Cerebrium.

Guide for the config files for fine-tuning a diffusion model on Cerebrium

Learn how to seamlessly deploy your trained model to a production environment.

Auto-Deploying your Trained Model

Stable Diffusion is an image generation model that can receive either text or image input and output network-generated images

Stable Diffusion

Image-to-Image text-guided generation with Stable Diffusion

Img2Img

Image-to-text model that specializes in producing prompts to work with Stable Diffusion 2

Img2text

Dreambooth is a fine-tuned text-to-image diffusion model for subject driven generation

Custom Dreambooth

A framework that allows for supporting various spatial contexts that can serve as additional conditionings to Diffusion models such as Stable Diffusion.

ControlNet

A Stable Diffusion Upscaler from Stability AI

Stable Diffusion Upscaler

A ESRGAN Upscaler built for photo restoration

ESRGAN Upscaler

A prompt based segmentation model from Meta AI

Segment Anything

Whisper is a language generation model that receives audio data and performs transcription and or translation, then outputs a text response.

Whisper

MT0 is a model capable of following human instructions in dozens of languages

A model used to answer questions or tasks across many languages

FlanT5

GPT-Neo is a transformer model designed using EleutherAI's replication of the GPT-3 architecture

GPT-Neo

Raw model for masked language modeling, but it's mostly intended to be fine-tuned on a downstream task

Home

Cortex

Conduit

Fine-tuning

Prebuilt Models

FAQs and Tips

Deploying Prebuilt Models