Deep-learning-in-cloud

zszaziMIT活跃维护

汇总各家提供深度学习 GPU 算力的云服务商清单,方便按需挑选训练和推理用的云端 GPU 资源。

List of Deep Learning Cloud Providers

暂无版本826star111fork同步于 2026-11-10 23:53 UTC

怎么上手

README 里没有找到可直接复制的安装命令

README 里没有找到可直接复制的安装命令,请到 README 或官方文档查看安装方式。

硬件要求按 README 说明

README 未说明;本文是云服务商清单,算力由所选云厂商提供,不涉及本地硬件要求

支持 ROCm / AMD
下载:源码 zip·GitHub Releases(安装包)本站不镜像安装包

项目说明

由 README 自动翻译提炼

一份汇总云端 GPU 训练、模型部署与 MLOps 平台的清单,帮你快速找到算力和上线渠道

汇总各家提供深度学习 GPU 算力的云服务商清单,方便按需挑选训练和推理用的云端 GPU 资源。

核心能力

  • 按云厂商列出 GPU 价格与免费额度
  • 单列模型 Web 部署平台对比表
  • 汇总 MLOps 全生命周期平台
  • 标注免费试用与赠送额度信息

容易踩的坑

这是资源清单而非可安装的软件,没有安装命令;价格与免费额度以各厂商官网为准,可能已变动;部分链接带推广参数

内容由模型根据 README 生成,可能不完整或不准确,以官方文档为准。看原文 README

最近发版

0 个版本

暂无 release 数据(可能没有发版,或尚未同步)。

用户点评

登录用户的真实使用记录,可按版本和硬件筛选
共 0 条

    还没有人点评。第一个说出真实体验的人最有价值。

    README 原文

    未翻译

    ☁️☁️Deep-learning-in-cloud☁️☁️

    Ever had a laptop 💻 which is not powerful enough to run your models , forget about it and use Cloud GPUs ☁️ to train your model faster and cheaper cuz its raining 💦 Cloud GPUs

    • Also get to know where you can deploy your model to serve millions of people

    • :new: List of MLOps Platforms for complete Machine Learning LifeCycle :recycle:

    • Check out the free credits and Perks/offer section to get some free gpu hours

    Cloud vendorWebsitePricingFree Trial / Free Credits
    Google Colaboratory :heart:https://colab.research.google.comFREEFREE FOREVER*
    Kaggle Kernelshttps://www.kaggle.comFREEFREE FOREVER*
    Ace Cloudacecloudhosting.compay as you go[$300 credits] (https://www.acecloudhosting.com/public-cloud/gpu/)
    Activeloophttps://www.activeloop.ai--
    Alibaba cloudhttps://www.alibabacloud.com/Pay as you go$300 credits
    AWS Sagemakerhttps://aws.amazon.com/sagemaker/pricing :label: Free plans
    Azurehttps://azure.microsoft.com/en-in/products/machine-learning/pricing :label: $200 credits
    Cirrascalehttps://www.cirrascale.compricing :label:-
    Cloudalize (now Radian Arc)radianarc.com--
    CloudGPUhttps://cloudgpu.app/?utm_source=github&utm_medium=referral&utm_campaign=202611_seg10_student&utm_content=gh_deep_learning_in_cloudpricing :label: per-minute billing (rounded up, 1-minute minimum), live per-GPU prices for RTX 3090, RTX 4090, RTX 5090, RTX A6000, A100 and RTX PRO 6000; OpenAI-compatible model API-
    DataCrunchhttps://verda.com/V100 at $0.69/h-
    Dataikuhttps://www.dataiku.com-Free Plans
    Dataoortshttps://dataoorts.com/pricing :label: Starting at $0.55/hr for 1x A100, $0.69/hr for 1x H100 and $0.88 for 1x GH200-
    Deepnotehttps://deepnote.com/Currently in Beta-
    Deploybasehttps://deploybase.ai/pricing-
    Exoscalehttps://www.exoscale.com/gpu/pricing :label: -
    Golemhttps://golem.network--
    Google Cloud Platformhttps://cloud.google.com/gpu/pricing :label: $300 credits
    gpu.nzhttps://gpu.nzper-minute :label: RTX 3090 from ~$0.32/hr, RTX 4090 from ~$0.45/hr, H100 from ~$2.69/hr; paid in rubles via SBP or crypto; one-click ComfyUI, Jupyter, Ollama, vLLM templates-
    GPUeaterhttps://gpueater.compricing :label: -
    GPULabhttps://gpulab.iopricing :label: -
    GrabGPUhttps://grabgpu.comFree tracker :label: daily prices and reported stock for 71 GPU models across 22 cloudsFree, no sign-up
    GridSharehttps://gridshare.inpricing :label: RTX 3060 from ₹12/hr, RTX 4090 ₹48/hr, RTX 5090 ₹51/hr, per-second INR billing₹25 trial credit on email verification
    Hinodehttps://hinode.runpricing :label: from $1.75/hr for 1x L4, $3.30/hr for 1x L40S and $8.05/hr for 1x RTX PRO 6000-
    Hostkeyhttps://hostkey.com/gpu-dedicated-servers/GPU from 90 euros/monthFree trials available
    Hundredth Run compute deskhttps://hundredthrun-compute-desk.pages.devFree price table :label: on-demand list prices for H100, H200, B200, A100, MI300X and more across 25 clouds, each with source link and date, updated weekly, CC BY 4.0 dataFree, no sign-up
    IBM Cloudhttps://www.ibm.com/products/gpu-ai-acceleratorPay as you go$200 credits
    Jarvis Labshttps://jarvislabs.ai/RTX 5000 at $0.49/hr- )
    Lambdahttps://lambda.ai/cloudStarting at $0.60/hr for a 1x A10 and $1.10/hr for a 1x A100-
    Lantern Workshttps://thelanternworks.com/?ref=deep-learning-in-cloudFree comparison :label: tracks on-demand GPU cloud rental prices across multiple providersFree, no sign-up
    Leadergpuhttps://www.leadergpu.compricing :label: -
    Liumhttps://lium.iopricing :label: per-second billing, live per-GPU prices for H100, H200, B200 and RTX 4090/5090-
    Lumino AIhttps://www.luminoai.inpricing :label: RTX 4090 from ~$1.10/hr, A100 80GB from ~$2.35/hr, per-second INR billing-
    Nimbleboxhttps://nimblebox.aipricing :label: Free $10 worth of cloud credits
    NoctHosthttps://nocthost.com/gpu-vpspricing :label: hourly billing, fractional NVIDIA A16 from $0.085/hr, A40 48 GB at $2.397/hr; pay by card or crypto-
    Nvidia cloudNvidia Cloud GPU--
    One Stop Systemhttps://onestopsystems.com/--
    Paperspacehttps://www.paperspace.compricing :label: Referal Program Available
    puzl.eehttps://puzl.cloud/gpu-cloudRent a fraction of A100 for 0.40EUR/hFree cloud Kubernetes API, up to 10 GPUs per pod
    QIThttps://qit.energypricing :label: L40s from $1.25/hr, H100 $3.08/hr, H200 $3.51/hr, B200 $4.68/hr; serverless + reserved; Model API$1.56 free credits for new business-email accounts
    Radeon Cloudhttps://radeon.anruicloud.com/?utm_source=github&utm_medium=deep-learning-in-cloud&utm_campaign=awesome-dlcloudPay as you goFree tier — ROCm GPU notebooks on AMD Radeon (Colab-style)
    Rapid Switchhttps://www.rapidswitch.compricing :label: -
    Runcratehttps://runcrate.ai/?utm_source=deep-learning-in-cloud&utm_medium=directory&utm_campaign=gpu_pricingpricing :label: per-second billing, live per-GPU prices for H200, H100, A100, L40S and RTX 4090/5090; 1-8 GPUs per instance; CLI and MCP server-
    RunPodhttps://www.runpod.io/pricing :label: Starting at $1.99/hr for 1x H100, $0.34/hr for 1x 4090, $3.99/hr for 1x H200Research & startup credits programs available
    Symbiozahttps://symbioza.devprepaid credit, from $10-
    TensorDockhttps://www.tensordock.com/pricing :label: Discounts to FOOS, students and researchers
    Vast.aihttps://vast.aipricing :label: -
    VoltageGPUhttps://voltagegpu.compricing :label: -

    Deploy your model as a Web app

    Have an idea and want to serve to world 🌎 , create a Webapp and deploy it as a flask , Django etc

    VendorWebsitePricingFree Trial / Free Credits
    Digital Oceanhttps://www.digitalocean.comPay as you goFree $100 credits with github student pack
    Glitchhttps://glitch.com--
    Herokuhttps://www.heroku.compricing :label: Free plan (model<500MB)
    PythonAnywherehttps://www.pythonanywhere.com/pricing :label: Free Beginner Account Available
    Renderhttps://render.compricing :label: -
    Streamlit For Teamshttps://streamlit.io/pricing :label: Currently in Beta ( Streamlit Cloud Tool )
    Vercelhttps://vercel.com/pricing :label: Free plan available

    MLOps Platforms

    A Beautiful marriage :ring: between Machine Learning and DevOps ( A Match Made in Heaven )

    Working on Serious Enterprise Level projects that has potential to serve millions of people and make :moneybag: , leave it to the power :zap: of DevOps to manage your Machine Learning LifeCycle

    Project / PlatformWebsitePricingFree Trial / Free Credits
    Akira.aihttps://www.akira.ai/pricing :label: -
    Algohttps://www.algomox.com/-Free Edition Available
    Algorithmiahttps://www.datarobot.com/product/ai-platform/pricing :label: -
    Amazon Sagemakerhttps://aws.amazon.com/sagemaker/pricing :label: Available for free as part of AWS Free Tier
    Arize AIhttps://arize.com/pricing :label: Free tier available
    Arriktohttps://www.arrikto.com/--
    ClearMLhttps://clear.mlpricing :label: Free plan available
    Cnvrghttps://www.intel.com/content/www/us/en/developer/tools/cnvrg/overview.htmlpricing :label: -
    DataRobothttps://www.datarobot.com/product/ai-platform/-$500 of free usage credits across products
    Fiddler AIhttps://www.fiddler.ai/pricing :label: SageMaker ML Monitoring With Fiddler AI :label:
    Flytehttps://flyte.org/-Open Source :octocat: Link
    Google Cloud AI Platformhttps://cloud.google.com/vertex-aipricing :label: -
    Gradient from Paperspacehttps://www.paperspace.com/artificial-intelligencepricing :label: -
    Hopsworkshttps://www.hopsworks.ai/pricing :label: Free plan available
    HPE - EzmeralSolution from HP-
    HPE - GreenLakeSolution from HP-
    Iguaziohttps://www.iguazio.com/mlops/-14 Day Free Trial
    KubeFlow ( for k8s )https://www.kubeflow.org/-Open Source :octocat: Link
    Lightning AI (formerly Grid.ai)https://lightning.ai/pricing :label: $25 free credits + special promo for researchers!
    MLFlowhttps://mlflow.org/-Open Source :octocat:
    Neu.rohttps://www.apolo.us/--
    Seldon Corehttps://seldon.io/--
    Valohaihttps://valohai.compricing :label: -

    Perks and offers

    If you are a student or researcher you can get extra credts , contact the provider

    • Paperspace provides $10 of free Gradient° credit fast.ai link

    • Radeon Cloud gives free AMD ROCm GPU notebook hours on AMD Radeon hardware (Colab-style, with ready-to-run templates for fine-tuning and inference) link

    • Do you have a GPU lying around rent your machine to Earn money using Vast.ai*

    • Test Drive Nvidia GPU link

    • AWS Cloud Credits for Research -link

    • Nvidia Academic Hardware Grant Program- link

    • If you are a Startup then google has you covered wth Startup Program giving you credits from $1000 to $100000 - link

    • Google giving cluster of 1000 TPUs to researcher In total, this cluster delivers a total of more than 180 petaflops of raw compute power! techcrunch link - application link

    • Google cloud Education Grant - link

    • Github Education pack - along with many offers has upto $110 credits for AWS - link

    • Watch out on fast.ai Forums to get coupon code for free credits

    • Want to use a Super Computer but don't have one, go for Golem - Golem is a decentralized marketplace for computing power. It enables CPUs and GPUs to connect in a peer-to-peer network, enabling both application owners and individual users to rent resources from other users machines, so turbo charge your next model training.

    • Hostkey provides grants for research, startups and competition winners link

    * Notes

    • Google colab and Kaggle kernels have limited session time

    • Most of the gpu providers run on top of AWS , GCP etc so may have more or less same pricing as the latter

    • Information given above is best to my searching ability , you may recheck with the provider for pricing and other info

    • license