# Cloud GPU

## GPU Rental

* [Vast.ai](https://console.vast.ai/create)
* [Lambda](https://lambdalabs.com/service/gpu-cloud)
* [CoreWeave](https://coreweave.com/gpu-cloud-pricing)
* [RunPod](https://www.runpod.io/gpu-instance/pricing)
* [Jarvislabs.ai](https://jarvislabs.ai/pricing/#nvidia-ampere-gpu)
* [DataCrunch](https://datacrunch.io/)
* [Fluidstack](https://console2.fluidstack.io/deploy)
* [Genesis Cloud](https://www.genesiscloud.com/pricing)
* [Crusoe Cloud](https://crusoecloud.com/pricing/)
* [Paperspace - Gradient Pricing](https://www.paperspace.com/gradient/pricing) Monthly

## Cloud Comparison

* [CloudOptimizer](https://cloudoptimizer.io/)
* [Cloud GPUs](https://cloud-gpus.com/)

## GPU Rent Price Comparison per hour

1xA100, 40GB, Vast, $0.828-$1.372\
1xA100, 40GB, Jarvis, $1.29\
1xA100, 40GB, Runpod, $0.89\
1xA100, 40GB, Lambda, $1.1\
1xA100, 40GB, Coreweave, $2.06\
1xA100, 40GB, CrusoeCloud, $2.35

1xA100, 80GB, Vast, $0.852\
1xA100, 80GB, Runpod, $2.04\
2xA100, 80GB, Lambda, $2.2\
1xA100, 80GB, Coreweave., $2.21\
1xA100, 80GB, Datacrunch, $2.2\
1xA100, 80GB, Fluidstack, $1.94\
1xA100, 80GB, CrusoeCloud, $2.85

1xA6000, 48GB, Vast, $0.587-$1.001\
1xA6000, 48GB, Jarivs, $0.99\
1xA6000, 48GB, Runpod, $0.74\
1xA6000, 48GB, Lambda, $0.80\
1xA6000, 48GB, Coreweave, $1.28\
1xA6000, 48GB, Datacrunch, $0.99

1xRTX3090, 24GB, Vast, $0.18-$0.353\
1xRTX3090, 24GB, Runpod, $0.39\
1xRTX3090, 24GB, Genesis, $1.30

1xV100, 16GB, Vast, $0.372\
1xV100, 16GB, Runpod, $0.28\
1xV100, 16GB, Coreweave, $0.80\
1xV100, 16GB, Google, $0.74\
1xV100, 16GB, Amazon, $0.918\
1xV100, 16GB, Datacrunch, $0.89\
1xV100, 16GB, Fluidstack. $0.8

## Storage Price per month

Runpod $0.20/GB\
Coreweave $0.04/GB

## Best Bang to Bucks GPU for ML

[GPU Benchmarks for Deep Learning - Lambda](https://lambdalabs.com/gpu-benchmarks)\
[AI-Benchmark](https://ai-benchmark.com/ranking_deeplearning)\
[The Best GPUs for Deep Learning in 2023 — An In-depth Analysis](https://timdettmers.com/2023/01/16/which-gpu-for-deep-learning/)\
[Razer Core X - Thunderbolt™ 3 eGPU](https://www.razer.com/gaming-egpus/razer-core-x) External GPU

* 12 GB RTX 3060 (recommended, cheap, low thermal, great memory) start from Rp6.000.000
* 12 GB A2000 (slightly less powerful than 3060 but only 75w power)
* 12 GB RTX 3060 Ti
* 24 GB RTX 3090 (two times performances of 3060) start form Rp12.000.000
* used Tesla V100 32GB

## GPU Rank

RTX 3090 > Tesla V100 32GB > (RTX 3080 = RTX 6000)

* 1.6: RTX A100 40GB > RTX 4090 > RTX A6000
* 1: RTX 3090 Ti 24 GB = V100 32 GB
* 0.8: RTX 3090 24GB = RTX 3080 Ti 12GB > RTX 3080
* 0.75-0.8: A6000 48GB = RTX 6000 24 GB = A4000
* 0.7 A40 = RTX 2080 Ti
* 0.6: RTX 3070 Ti
* 0.5: GTX 1080 Ti, RTX 2070, RTX 2080
* Tesla P100 > P6000
* 0.45-0.48: RTX 3070, RTX 3060 Ti
* 0.35: GTX 1080, RTX 3060, RTX 3050
* 0.25: GTX 1070

A100 > V100 > T4

## Machine Learning Deployment

* [DataCrunch - A100 80GB GPU Servers](https://datacrunch.io/)
* [Towhee - Out-of-box Pipelines - Towhee](https://towhee.io/pipelines?limit=30\&page=1)
* [Rent GPU Servers for Deep Learning and AI - Vast.ai](https://vast.ai/)
* [Replicate–Reproducible machine learning](https://replicate.com/)
* [Gradient - Use machine learning to make anything.](https://gradient.run/)
* [AutoGluon: AutoML for Text, Image, and Tabular Data—AutoGluon Documentation 0.4.0 documentation](https://auto.gluon.ai/stable/index.html)
* [Parsec–Help Center Home](https://support.paperspace.com/hc/en-us/articles/115002289833-Parsec)
* [codeplea/genann: simple neural network library in ANSI C](https://github.com/codeplea/genann)
* [DerekChia/colab-vscode: ✨ 1-Click Free GPU on VS Code with Google Colab](https://github.com/DerekChia/colab-vscode)
* [Code examples](https://keras.io/examples/)
* [Theory of Machine Learning](https://www.tml.cs.uni-tuebingen.de/teaching/2020_maths_for_ml/)
* [Home - Deploifai](https://deploif.ai/)
* [ELBO AI - ML training made easy and cost effective](https://www.elbo.ai/)


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://irosyadi.gitbook.io/irosyadi/app/cloud-gpu.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.