Brev.dev, inc of San Francisco, California, USA has been acquired by NVIDIA Corporation of Santa Clara, California, USA on July 2024

Learn more
BlogPricing

Easy-to-use GPUs.

Fine-tune, train, or deploy. Use your own notebook, or one of ours. SSH too. CUDA, Python, Jupyter Lab, all set up.

Read the docs

Fine-tune Llama3-8B

Build your own LLM by fine-tuning Llama3-8B on your own data.

Deploy LLaVa with 1-click

Fine-tune the LLaVa model on your images to describe them.

Run SDXL using ComfyUI and TensorRT

Use OCR + MistralLite to chat with your PDFs.

Aman
@_amankishore
@brevdev x @BananaDev_ is too OP of a setup 😍
Daniel Whitenack
@dwhitena
I had a delightful time setting up a #GPU-based dev environment... When's the last time you said that? 😀
John Simerlink
@johnsimerlink
@brevdev I just had the best developer experience and customer service I may have had with any product ever. I had spent hours trying to rent GPUs and configuring environments correctly. I stumbled across brev and 10 minutes later had it running and working locally
Aman
@_amankishore
Been using @brevdev for 5 min today and I’m sold! Brev makes it ridiculously easy to spin up spot instances for development (60% cheaper than on-demand), excited to save a lot more on dev costs this month!
Mitesh
@mitesh711
@brevdev's offering is very smooth. Don't have to care about underlying GPU/cloud at all.
Rony
@ronyfadel
Brev saved me so much time developing and deploying ML apps, compared to running jobs locally. With Brev, I don’t have to go through clunky AWS or GCP UI’s, spend an hour setting up an instance just to realize that GPU’s are not available in that region.
Jasper
@Jasperschoormns
Really recommend @brevdev for GPU programing. Your own powerful GPU server, but its cheap and so easy to set up. Couldn't do it locally

Official Partners

Trusted By Engineers From

Radically Better Developer Experience

Easy-to-Use GPUs

Use Brev.dev to find a GPU and get it configured to fine-tune or train your model.

Get a GPU (or CPU).
A single interface between AWS, GCP, Fluidstack, and other GPU clouds. Use credits when you have them. Choose an instance based on costs & availability.
Set Up Software.
CUDA just works. Dependencies installed properly. No more Python environment issues. Sit back and let us do the work.
Effortless Access
Access your instance from anywhere via Jupyter Notebook, VSCode, or through our CLI. Simply run brev open instance-name and get started.

Getting started

Welcome to Brev.dev

Brev is a AI/ML development platform that allows you to run build, train, and deploy ML models on the cloud. Brev allows you start small on a CPU instance and effortlessly scale to larger GPU clusters for any workload.

Why should I use Brev?

  • Build for devs: Brev is built for developer by developers. We understand the challenges of building and scaling ML projects and built Brev to a lightweight but powerful platform for any type of workload.
  • Cost-effective: Brev aggregates multiple cloud providers to find the right type of GPU for best price possible.
  • Powerful: Brev allows you to run any code on your GPU instance. You can use Brev to train and deploy ML models on the cloud.

Getting help

We want to create as many channels for help as possible; if you'd like a new channel, please let us know!

Join the community

The best way to reach us is via our Discord! Go to the #help channel and create a support ticket with /help command.

Contact us

[email protected]

Shared Slack channel

Paid users get a prioritized shared Slack channel with our team!

Next
Overview