Instagram Linkedin X-twitter Youtube
Linkedin
X-twitter
Facebook
runpod-logo

RunPod

Get in Touch
Get a Free Expert Consultation
Compute AI & GPU Infrastructure Developer Tools & Ecosystem

Quick Stats

Pricing Model
Pay-As-You-Go, Reserved / Committed Capacity, Spot Pricing
Customer Count
500k
Region
US West, US East, US Central, CA East, EU West, EU Central, EU North, EU East, Oceania (AU)

About

RunPod offers on-demand GPU cloud infrastructure for AI, machine learning, and compute-intensive workloads. Its platform features pay-as-you-go pricing, instant deployment, and compatibility with popular ML frameworks, making it ideal for developers seeking flexible, cost-effective access to high-performance GPUs without long-term commitments.

Explore
Alteranate Cloud Providers
backblaze-logo
Backblaze
Wasabi-Logo-600x250-1 3
Wasabi
Ionos
crunchydata-logo
Crunchy Data
  • Newsletter

Stay Ahead in Cloud Infrastructure

Join the community building scalable, cost-efficient cloud strategies with insights from experts

Subscribe to Newsletter
Pricing Details

RunPod Pricing

Cloud GPUs, Serverless AI, Clusters, and Storage
Cloud GPU Pods
On-demand GPUs in 30+ regions
Per-second billing
GPU Examples
H200 (141GB VRAM): Contact sales
B200 (180GB VRAM): Contact sales
H100 SXM (80GB VRAM): Contact sales
A100 SXM (80GB VRAM): $1.79/hr
L40S (48GB VRAM): Contact sales
Many more GPU types available
Billing Model
• Per-second and per-hour billing
• No minimum commitment
Pricing varies by GPU, region, and availability. See full list on RunPod's pricing page.
Serverless GPU Endpoints
Instant AI workloads, no idle costs
Flex & Active workers
GPU Example
B200 (180GB): $8.64/s (Flex), $6.84/s (Active)
H200 (141GB): $5.58/s (Flex), $4.46/s (Active)
H100 (80GB): $4.18/s (Flex), $3.35/s (Active)
A100 (80GB): $2.72/s (Flex), $2.17/s (Active)
L40/L40S/6000 Ada (48GB): $1.9/s (Flex), $1.33/s (Active)
Many more GPU types available
Worker Types
• Flex: Scales up during spikes, cost-efficient for bursty workloads
• Active: Always-on, up to 30% discount, eliminates cold starts
Save 25%+ over other serverless providers on flex workers.
Instant & Reserved Clusters
Multi-GPU clusters, scale up to 64 GPUs
Contact sales for pricing
Instant Clusters
H200 SXM: $4.31/hr
A100 SXM: $1.79/hr
Other GPUs: Contact sales
Reserved Clusters
Dedicated clusters, custom configs, SLA-backed uptime
Discounted rates for 10,000+ GPUs
Contact sales for all reserved cluster pricing
Reserved clusters require a minimum commitment. Pricing varies by configuration and term.
Storage
Persistent and network storage
No ingress/egress fees
Container Disk $0.10/GB/mo
Volume Disk
Running: $0.10/GB/mo
Idle: $0.20/GB/mo
Network Storage (Standard)
Under 1TB: $0.07/GB/mo
Over 1TB: $0.05/GB/mo
Network Storage (High-Performance)
Under 1TB: $0.14/GB/mo
Over 1TB: $0.07/GB/mo
No fees for ingress/egress. Persistent and temporary storage available.
Public AI Endpoints
Pre-deployed models via API
Pay-per-use
Audio Models
Pruna / Whisper V3 Large: $0.05 per 1,000 characters
resembleai / Chatterbox Turbo: $0.00 per 1,000 characters
Image Models
bytedance / Seedream 4.0 Edit: $0.0270 per request
pruna / Pruna Image T2I: $0.0050 per request
Language Models
deep-cogito / Deep Cogito v2 Llama 70B: $0.00001 per 1M tokens
ibm / IBM Granite 4.0 H Small: $1.00 per 1M tokens
Video Models
Bytedance / Seedance 1.0 pro: 5s $0.12 (480p) per request
Alibaba / Wan 2.2 I2V 720p: 5s $0.30 per request
Full list of models and pricing available on RunPod's public endpoints page.

Notes: Prices shown in USD. Actual rates may vary by GPU, region, and availability. Some high-end GPU and cluster pricing is available only upon request. See RunPod's official pricing page for the latest details.

Free Tool

Free Cost Calculator

Deep Dive on your Cloud Storage Costs

Calculate my Costs
Features
Compute On-demand and serverless GPU compute with support for over 30 GPU SKUs, including B200s and RTX 4090s.
Storage Persistent S3-compatible network storage with zero ingress/egress fees for full AI pipelines.
Performance Autoscaling from 0 to 1000s of GPU workers in seconds with sub-200ms cold-starts using FlashBoot.
Compliance SOC 2 Type II certified for end-to-end data protection and enterprise-grade uptime.
Key Offerings
 
  
       
  • On-demand Cloud GPUs across 30+ SKUs and global regions
    •    
  • Serverless AI workloads with autoscaling and sub-200ms cold starts
    •    
  • Multi-node GPU Clusters for large-scale training and deployment
    •    
  • Persistent S3-compatible network storage with zero egress fees
    •    
  • RunPod Hub for rapid deployment of open-source AI models
    •   
 
Ideal Use Cases / Buyers
Workloads AI/ML model training, real-time inference, fine-tuning, deployment of AI agents, compute-heavy tasks such as data processing and simulation, and scalable GPU-based workloads.
Buyers AI startups, ML engineers, generative AI platforms, SaaS companies, media & entertainment, architectural visualization, and developer teams at technology companies.
Integrations & Partners
Why Choose

     
  • ✅ Instantly deploy GPU-enabled environments in under a minute, supporting over 30 GPU SKUs across 8+ global regions
    •  
  • ✅ Serverless infrastructure that auto-scales from 0 to thousands of GPU workers in seconds, with sub-200ms cold-starts and zero idle costs
    •  
  • ✅ S3-compatible persistent storage with zero ingress/egress fees, enabling full AI pipelines without additional data transfer costs
    •  
  • ✅ Enterprise-grade reliability with 99.9% uptime and independently audited SOC 2 Type II compliance for data security
    •  
  • ✅ Proven cost efficiency, with customer case studies reporting up to 90% savings on infrastructure bills and the ability to match demand without overpaying for idle resources

Newsletter

Stay Ahead in Cloud
& Data Infrastructure

Get early access to new tools, insights, and research shaping the next wave of cloud and storage innovation.
Talk to expert

Get a free deep dive with an expert​

runpod-logo
Datastorage Expert
Analyst Report

Please enter your email to get this report now and on your email.