RunPod

Skip to main content4.8 stars750,000+ developersAI infrastructure developers trustEverything you need to train, deploy, and scale AI all in one place.Thank you! Your submission has been received!Oops! Something went wrong while submitting the form.Trusted by more than 750,000+ developers at the world’s leading AI companiesOwner NameOwner NameOwner NameOwner NameWhat’s newOpenAI's Parameter Golf: Train the Best Language Model That Fits in 16MB on RunpodMarch 25, 2026OpenAI just launched one of the most exciting ML competitions in years, and Runpod is the official compute partner. Here's everything you need to know to get started. Runpod named OpenAI's infrastructure partner for the Model Craft Challenge SeriesMarch 18, 2026Runpod and OpenAI will distribute up to $1M in compute credits supporting the first challenge, Parameter Golf.State of AI infrastructure reportInsights from the latest data around AI deployment, infrastructure demand, and model scaling trendsDownload the report -> SolutionRunpod makes GPU infrastructure simple.Runpod is the end-to-end AI cloud that
simplifies building and deploying models.Launch a GPU pod in seconds. Spin up a fully-loaded, GPU-enabled environment in under a minute. From B200s to RTX 4090s, Runpod supports over 30 GPU SKUs. Deploy globally with a few clicks.Run workloads across 8+ regions worldwide, with low-latency performance and global reliability.Scale on autopilot with Serverless.From 0 to 100 compute workers, adapting to your workload in real-time—only pay for what you use. Go from idea to deployment in a single flow.‍Runpod simplifies every step of your workflow—so you can build, scale, and optimize without ever managing infrastructure.‍‍Get started ->Spin upGo from idea to execution in seconds
—no provisioning, no delays.BuildTrain models, render simulations, or process data—without limits or lock-ins.IterateExperiment confidently with instant feedback and safe rollbacks.DeployAuto-scale across regions—zero idle costs, zero downtime.Enterprise grade uptime.Runpod handles failovers, ensuring your workloads run smoothly—even when resources don’t.Managed orchestration.Runpod Serverless queues and distributes tasks seamlessly, saving you from building orchestration systems.Real-time logs.Get real-time logs, monitoring, and metrics—no custom frameworks required.FeaturesScale with Serverless when you're ready for production. Powerful compute, effortless deployment.Try Serverless ->Autoscale in secondsInstantly respond to demand with GPU workers that scale from 0 to 1000s in seconds.Learn about autoscalingZero cold-starts with active workersAlways-on GPUs for uninterrupted execution.Learn about always-on<200ms cold-start with FlashBootLightning-fast scaling with
sub-200ms cold-starts.Discover FlashBootPersistent network storageRun full AI pipelines—data ingestion to deployment—without egress fees on our S3 compatible storage.Learn about storageCase StudiesLoved by developers.But don’t just take it from us. Play videoRunpod’s scalable GPU infrastructure gave us the flexibility we needed to match customer traffic and model complexity—without overpaying for idle resources.—Read case study Play video"Runpod has changed the way we ship because we no longer have to wonder if we have access to GPUs. We've saved probably 90% on our infrastructure bill, mainly because we can use bursty compute whenever we need it."—Read case studyhttps://media.getrunpod.io/latest/aneta-video-1.mp4 Play video"Runpod has allowed the team to focus more on the features that are core to our product and that are within our skill set, rather than spending time focusing on infrastructure, which can sometimes be a bit of a distraction.”—Read case studyhttps://media.getrunpod.io/latest/gendo-video.mp4 Play video"Runpod helped us scale the part of our platform that drives creation. That’s what fuels the rest—image generation, sharing, remixing. It starts with training."—Read case study Play video"Runpod allowed us to reliably handle scaling from zero to over 1,000 requests per second in our live application."—Read case studyhttps://media.getrunpod.io/latest/scatter-lab-video.mp4 Play video"Runpod has allowed us to focus entirely on growth and product development without us having to worry about the GPU infrastructure at all."—Bharat, Co-founder of InstaHeadshotsRead case studyhttps://media.getrunpod.io/latest/magic-studios-video.mp4 Play video"We could stop worrying about infrastructure and go back to building. That’s the real win.”—Read case study Play video“The main value proposition for us was the flexibility Runpod offered. We were able to scale up effortlessly to meet the demand at launch.”—Josh Payne, Coframe CEORead case study Play video"After migration, we were able to cut down our server costs from thousands of dollars per day to only hundreds."—Read case study Play videoRunpod’s scalable GPU infrastructure gave us the flexibility we needed to match customer traffic and model complexity—without overpaying for idle resources.—Read case study Play video"Runpod has changed the way we ship because we no longer have to wonder if we have access to GPUs. We've saved probably 90% on our infrastructure bill, mainly because we can use bursty compute whenever we need it."—Read case studyhttps://media.getrunpod.io/latest/aneta-video-1.mp4 ImpactGet more done for every dollar.More throughput, faster scaling, and higher efficiency—with Runpod, every dollar works harder.Get startedSee pricing ->Runpod175,301 tokensAzure67,559 tokensGCP42,637 tokensAWS38,370 tokensThis graphic shows tokens per dollar>500 millionServerless requests monthly57%Average reduction in setup timeUnlimitedData processed with zero ingress/egress feesEnterprise gradeEnterprise-grade from day oneBuilt for scale, secured for trust, and designed to meet your most demanding needs.Get startedTalk to a cloud specialist 99.9% UptimeRun critical workloads with confidence, backed by industry-leading reliability.Secure by defaultIndependently audited SOC 2 Type II compliance for end-to-end data protection.Scale to thousands of GPUs Adapt instantly to demand with infrastructure that grows with you.Build what’s next.The most cost-effective platform for building, training, and scaling machine learning models—ready when you are.Get startedRequest a demo You’ve unlocked areferral bonus!Sign up today and you’ll get a random credit bonus between $5 and $500 when you spend your first $10 on Runpod.Claim Your BonusWe use cookies to provide the content and functionality of this websiteAcceptRejectCookie PreferencesPersonalization CookiesMarketing CookiesAnalytics CookiesThank you! Your submission has been received!Oops! Something went wrong while submitting the form.CancelSave --- PodsThousands of GPUs across 30+ regions. Simple pricing plans for teams of all sizes,designed to scale with you.Get started ->GPUCommunity CloudSecure CloudPer secondPer hour>80GB VRAMH200141GB VRAM276GB RAM24vCPUs$3.59/hr‍B200180GB VRAM283GB RAM28vCPUs$4.99/hr‍RTX Pro 600096GB VRAM188GB RAM16vCPUs$1.89/hr‍H100 NVL94GB VRAM94GB RAM16vCPUs$3.07/hr‍80GB VRAMH100 PCIe80GB VRAM188GB RAM16vCPUs$2.39/hr‍H100 SXM80GB VRAM125GB RAM20vCPUs$2.69/hr‍A100 PCIe80GB VRAM117GB RAM8vCPUs$1.39/hr‍A100 SXM80GB VRAM125GB RAM16vCPUs$1.49/hr‍48GB VRAML40S48GB VRAM94GB RAM16vCPUs$0.86/hr‍RTX 6000 Ada48GB VRAM167GB RAM10vCPUs$0.77/hr‍A4048GB VRAM50GB RAM9vCPUs$0.40/hr‍L4048GB VRAM94GB RAM8vCPUs$0.99/hr‍RTX A600048GB VRAM50GB RAM9vCPUs$0.49/hr‍32GB VRAMRTX 509032GB VRAM35GB RAM9vCPUs$0.89/hr‍24GB VRAML424GB VRAM50GB RAM12vCPUs$0.39/hr‍RTX 309024GB VRAM125GB RAM16vCPUs$0.46/hr‍RTX 409024GB VRAM41GB RAM6vCPUs$0.59/hr‍RTX A500024GB VRAM25GB RAM9vCPUs$0.27/hr‍See all GPUsServerlessCost effective for every inference workload. Save 25% over other Serverless cloud providers on flex workers alone.Get started -> GPUPer secondPer hourFlexWorkers that scale up during traffic spikes and return to idle after completing jobs. Cost-efficient and ideal for bursty workloads.ActiveAlways-on workers that eliminate cold starts. Billed continuously but come with up to 30% discount.$0.00240/s$0.00190/s180GBB200Maximum throughput for big models.$0.00155/s$0.00124/s141GBH200Extreme throughput for big models.$0.00116/s$0.00093/s80GBH100PROExtreme throughput for big models.$0.00076/s$0.00060/s80GBA100High throughput GPU, yet still very cost-effective.$0.00053/s$0.00037/s48GBL40, L40S, 6000 AdaPROExtreme inference throughput on LLMs like Llama 3 7B.$0.00034/s$0.00024/s48GBA6000, A40A cost-effective option for running big models.$0.00044/s$0.00031/s32GB5090PROExtreme throughput for small-to-medium models.$0.00031/s$0.00021/s24GB4090PROExtreme throughput for small-to-medium models.$0.00019/s$0.00013/s24GBL4, A5000, 3090Great for small-to-medium sized inference workloads.$0.00016/s$0.00011/s16GBA4000, A4500, RTX 4000, RTX 2000The most cost-effective for small models.Instant ClustersLaunch multi-GPU clusters in minutes with no commitments—scale up to 64 GPUs, attach shared storage, and pay only for what you use.Get started ->GPUPer secondPer hourH200 SXMContact sales$4.31/hrA100 SXMContact sales$1.79/hrH100 SXMContact sales$/hrL40SContact sales$/hrB200Contact sales$/hrReserved ClustersDedicated GPU clusters with guaranteed availability, custom configurations, SLA-backed uptime, and discounted rates for enterprises scaling to 10,000+ GPUs.Talk to an engineerGPU1mo3mo6mo12mo12mo+H200 SXMContact salesA100 SXMContact salesH100 SXMContact salesL40SContact salesB200Contact salesStorageFlexible and persisitent storage options starting at $0.05/GB/mo with standard and high-performance tiers.Get started ->Storage TypeContainer Disk$0.10/GB/moVolume DiskRunning - $0.10/GB/moIdle - $0.20/GB/moNetwork Storage (Standard)Under 1TB - $0.07/GB/moOver 1TB - $0.05/GB/moNetwork Storage (High-Performance) $0.14/GB/moPublic EndpointsInstant access to pre-deployed AI models via API—no infrastructure setup required.Get started ->Model NameAudioPruna / Whisper V3 Large$0.05 per 1000 charactersresembleai / Chatterbox Turbo $0.00 per 1000 characters.minimax / Minimax Speech 02 HD $0.05 per 1000 charactersminimax / Minimax Speech 02 HD$0.05 per 1000 charactersImagebytedance / Seedream 4.0 Edit$0.0270 per requestbytedance / Seedream 4.0 T2I$0.0270 per requestgoogle / Nano Banana Edit$0.0380 per requestgoogle / Nano Banana Pro Edit$0.14 per requestpruna / Pruna Image T2I$0.0050 per requestpruna / Pruna Image Edit$0.01 per requestalibaba / WAN 2.6 T2I$0.03 per requestqwen / Qwen Image Edit 2511$0.02 per requestqwen / Qwen Image Edit 2511 LoRA$0.025 per requestTongyi-MAI / Z Image Turbo$0.0050 per request.Languagedeep-cogito / Deep Cogito v2 Llama 70B$0.00001 per 1m tokensqwen / Qwen3 32B AWQ$10.00 per 1m tokensminimax / Minimax Speech 02 HD $0.05 per 1000 charactersminimax / Minimax Speech 02 HD$0.05 per 1000 charactersibm / IBM Granite 4.0 H Small$1.00 per 1m tokensVideoBytedance / Seedance 1.0 pro5s: $0.12(480p) per requestAlibaba / Wan 2.2 I2V 720p5s: $0.30 per requestAlibaba / Wan 2.2 T2V 720p5s: $0.30 per requestAlibaba / Wan 2.1 I2V 720p$0.30 per requestAlibaba / Wan 2.1 T2V 720p$0.30 per requestkwaivgi / Kling v2.6 Standard Motion Control1-3s $0.21 per requestAlibaba / WAN 2.6 T2V5s: $0.50 per requestbytedance / Seedance V1.5 Pro I2V$0.024 per secondkwaivgi / Kling Video O1 R2V$0.112 per secondAlibaba / Wan 2.6 I2V5s: $0.50 per requestStorage PricingFlexible, cost-effective storage for every workload.No fees for ingress/egress. Persistent and temporary storage available.Pod PricingStorage TypeRunning PodsIdle PodsVolume$0.10/GB/mo$0.20/GB/moContainer Disk$0.10/GB/moNAPersistent Network StorageStorage TypeUnder 1TBOver 1TBNetwork Volume$0.07/GB/mo$0.05/GB/moGain additional savings
with reservations.Save more with long-term commitments. Speak with our team to reserve discounted active and flex workers.Get startedBuild what’s next.The most cost-effective platform for building, training, and scaling machine learning models—ready when you are.Get startedRequest a demo You’ve unlocked areferral bonus!Sign up today and you’ll get a random credit bonus between $5 and $500 when you spend your first $10 on Runpod.Claim Your BonusWe use cookies to provide the content and functionality of this websiteAcceptRejectCookie PreferencesPersonalization CookiesMarketing CookiesAnalytics CookiesThank you! Your submission has been received!Oops! Something went wrong while submitting the form.CancelSave --- AboutBuilt by developers,for developers.Join the team of people redefining the future of cloud computingSome of the most important companies of this decade will be built on Runpod.We aim to create the foundational platform for developers to build and run custom AI systems that scale.Give a sh*tWe work with people who care—about our customers and each other.Look in the mirrorWe deeply reflect on our own actions and seek to better ourselves.Choose courage over comfortWe tackle hard truths and tough situations directly—even when it makes us uncomfortable.Our guiding virtues.These values don't just define us; they are us. As we evolve, we will continue to let our brand reflect who we truly are, growing richer over time and earning deeper value and trust.Unlike our products, which will adapt and change, our brand's core identity will remain consistent, changing only with deliberate effort and broad consensus.Careers at Runpod.If our vision speaks to you, and you’re passionate about building tools that push the boundaries of AI, come join us.See open rolesWe wear multiple hats.Our sales team has a strong technical background. Our eng team has an exceptional eye for product. Our datacenter operations team has a strong sales background. Whether you're doing engineering, sales, operations, design, or product, you should be comfortable wearing more than one hat at a time.We are customer-obsessed.We have over 300,000 developers that rely on us to run their workloads. Many of them use Runpod for their production environments. That's a huge responsibility that we don't take lightly. We are constantly talking to our customers to understand what we can do to level up our platform.We contribute cross-functionally.If you think there's a better way of doing X, even if it doesn't directly fall under your function, you should voice that to the team. The most important products we've built have stemmed from conversations between non-overlapping teams, like ML + Sales, Customer Support + Product, and Datacenter Operations + Accounting.We are agile.At this stage, we need to move quickly. The faster we can ship products that delight our customers, the more successful they will be with our platform. The AI industry is constantly evolving, and we must be able to evolve just as quickly. There are some things we believe won't change. Developers will always want faster, more accessible and cost-effective compute. Although our products may change, our mission will stay the same.We’re looking for people who want to grow with Runpod.We’re growing fast and building toward a team of several hundred in the years ahead. We’re looking for people who don’t just execute but lead. If you have a bias toward ownership, adaptability, and helping us scale, you’ll fit right in.Remote and hybrid.We’re a remote-first, globally distributed team spanning the U.S., Canada, Europe, and India. While most of us work remotely, we also maintain a presence in San Francisco, where part of the team is based and where we regularly host events.Join usBuild what’s next.The most cost-effective platform for building, training, and scaling machine learning models—ready when you are.Get startedRequest a demo You’ve unlocked areferral bonus!Sign up today and you’ll get a random credit bonus between $5 and $500 when you spend your first $10 on Runpod.Claim Your BonusWe use cookies to provide the content and functionality of this websiteAcceptRejectCookie PreferencesPersonalization CookiesMarketing CookiesAnalytics CookiesThank you! Your submission has been received!Oops! Something went wrong while submitting the form.CancelSave --- Cloud GPUsHigh-performance GPUs on demand.Run AI, ML, and HPC workloads on powerful cloud GPUs—without limits or wasted spend.Get started ->Blink and it’s ready.Deploy GPUs in under a minute—no need to wait for provisioning.Scale globally.Spin up one or hundreds of GPUs across 31 regions.Pay by the second.Ultra-flexible, on-demand billing—no commitments. GPU PricingThousands of GPUs across 30+ regions.Simple pricing plans for teams of all sizes, designed to scale with you.Get startedTalk to a cloud specialistGPUCommunity CloudSecure CloudPer secondPer hour>80GB VRAMH200141GB VRAM276GB RAM24vCPUs$3.59/hr‍B200180GB VRAM283GB RAM28vCPUs$5.98/hr‍RTX Pro 600096GB VRAM188GB RAM16vCPUs$1.69/hr‍H100 NVL94GB VRAM94GB RAM16vCPUs$2.59/hr‍80GB VRAMH100 PCIe80GB VRAM188GB RAM16vCPUs$1.99/hr‍H100 SXM80GB VRAM125GB RAM20vCPUs$2.69/hr‍A100 PCIe80GB VRAM117GB RAM8vCPUs$1.19/hr‍A100 SXM80GB VRAM125GB RAM16vCPUs$1.39/hr‍48GB VRAML40S48GB VRAM94GB RAM16vCPUs$0.79/hr‍RTX 6000 Ada48GB VRAM167GB RAM10vCPUs$0.74/hr‍A4048GB VRAM50GB RAM9vCPUs$0.35/hr‍L4048GB VRAM94GB RAM8vCPUs$0.69/hr‍RTX A600048GB VRAM50GB RAM9vCPUs$0.33/hr‍32GB VRAMRTX 509032GB VRAM35GB RAM9vCPUs$0.69/hr‍24GB VRAML424GB VRAM50GB RAM12vCPUs$0.44/hr‍RTX 309024GB VRAM125GB RAM16vCPUs$0.22/hr‍RTX 409024GB VRAM41GB RAM6vCPUs$0.34/hr‍RTX A500024GB VRAM25GB RAM9vCPUs$0.16/hr‍See all GPUs"The Runpod team has clearly prioritized the developer experience to create an elegant solution that enables individuals to rapidly develop custom AI apps or integrations while also paving the way for organizations to truly deliver on the promise of AI."Amjad Masad"Runpod is the only place I can deploy high-end GPU models instantly—no sales calls, no rate limits, no nonsense."Daniel Chang“The main value proposition for us was the flexibility Runpod offered. We were able to scale up effortlessly to meet the demand at launch.”Josh Payne“Runpod helped us scale the part of our platform that drives creation. That’s what fuels the rest—image generation, sharing, remixing. It starts with training.”Matty ShimuraAI AppsResearchLLM InferenceModel TrainingDeveloper ToolsBuilt-in developer tools & integrations.Powerful APIs, CLI, and integrations
that fit right into your workflow.Get startedTalk to a cloud specialistFull API access.Automate everything with a simple, flexible API.CLI & SDKs.Deploy and manage directly from your terminal.GitHub & CI/CD.Push to main, trigger builds, and deploy in seconds.Storage PricingFlexible, cost-effective storage for every workload.No fees for ingress/egress. Persistent and temporary storage available.Pod PricingStorage TypeRunning PodsIdle PodsVolume$0.10/GB/mo$0.20/GB/moContainer Disk$0.10/GB/moNAPersistent Network StorageStorage TypeUnder 1TBOver 1TBNetwork Volume$0.07/GB/mo$0.05/GB/moGain additional savings
with reservations.Save more with long-term commitments. Speak with our team to reserve discounted active and flex workers.Get startedFAQsQuestions? Answers.Curious about unlocking GPU power in the cloud? Get clear answers to accelerate your projects with on-demand high-performance compute.What are GPU Pods and how do they differ from other cloud GPU offerings?GPU Pods are dedicated GPU instances you can spin up on Runpod. Unlike abstracted serverless GPUs, Pods give you full control over the underlying VM, drivers, and environment. You get a persistent instance (or ephemeral, if you prefer) with direct access to powerful GPUs, letting you run training, inference, or other workloads exactly how you want.Which GPU models are available?We offer 30+ GPU models, from entry-level inference cards to top-tier training accelerators. Examples include A100, H100, RTX 6000 Ada, L4/L40 series, and many more—over 30 options in total. You can pick any supported GPU when you launch a Pod, and new models roll out as soon as they’re live on the platform. For the latest availability, check the dashboard or query the API.How is pricing structured?Pricing is shown as an hourly rate but billed by the millisecond. You only pay for the exact time your Pod runs—if you start and stop a Pod in one minute, you’re charged just that minute. Storage volumes may incur minimal fees when attached, but compute costs are metered by the millisecond.Can I bring my own Docker container or environment?Yes. GPU Pods support custom Docker images. You can build an image with your preferred libraries and push it to a registry (Docker Hub, ECR, etc.), then reference it when you launch the Pod. That way you control the OS, drivers, and dependencies.Which frameworks and runtimes are supported?Any framework that runs on Linux and supports GPUs: PyTorch, TensorFlow, JAX, ONNX, CUDA toolkits, etc. Since you control the container, you can install whatever versions or additional tools you need (e.g., NCCL, Horovod). We provide base images with common ML stacks to speed up setup.What about spot/preemptible GPUs?We offer spot instances where GPU capacity is available at a discount, but with the risk of eviction when demand spikes. You can use them for fault-tolerant or batch workloads. The UI/API will indicate current spot availability and pricing. ClientsTrusted by today's leaders, built for tomorrow's pioneers.Engineered for teams building the future.Build what’s next.The most cost-effective platform for building, training, and scaling machine learning models—ready when you are.Get startedRequest a demo You’ve unlocked areferral bonus!Sign up today and you’ll get a random credit bonus between $5 and $500 when you spend your first $10 on Runpod.Claim Your BonusWe use cookies to provide the content and functionality of this websiteAcceptRejectCookie PreferencesPersonalization CookiesMarketing CookiesAnalytics CookiesThank you! Your submission has been received!Oops! Something went wrong while submitting the form.CancelSave