Powering the next generation of AI & high-performance computing.
Engineered for large-scale AI training, deep learning, and high-performance workloads, delivering unprecedented compute power and efficiency.
NVIDIA Hopper Architecture
Breakthrough architecture designed for transformer models delivering up to 30X faster LLM inference performance.
Fourth-Generation Tensor Cores
Advanced AI acceleration with Transformer Engine and FP8 precision delivering up to 9X faster training.
80GB HBM3 Memory
High-bandwidth memory with 3.35TB/s bandwidth enables training and inference on the largest AI models.
NVLink Connectivity
High-speed GPU-to-GPU interconnect at 900GB/s enables seamless scaling across multiple GPUs for distributed training and massive workloads.
Performance
Key specs at a glance.
Performance benchmarks that push AI, ML, and HPC workloads further.
Memory Bandwidth
3.35
TB/s
FP16 Tensor Performance
1.98
PFLOPS
NVLink Bandwidth
900
GB/s
Use Cases
Popular use cases.
Designed for demanding workloads
—learn if this GPU fits your needs.
Technical Specs
Ready for your most demanding workloads.
Essential technical specifications to help you choose the right GPU for your workload.
Specification
Details
Great for...
Memory Bandwidth
3.35
TB/s
Feeding large model weights and data into HBM3 without stalls—crucial for trillion-parameter LLMs and HPC simulations.
Memory Bandwidth
1.98
Feeding large model weights and data into HBM3 without stalls—crucial for trillion-parameter LLMs and HPC simulations.
Feeding large model weights and data into HBM3 without stalls—crucial for trillion-parameter LLMs and HPC simulations.
3.35
FP16 Tensor Performance
1.98
PFLOPS
Accelerating mixed-precision training and inference on massive transformer and HPC models.
FP16 Tensor Performance
1.98
Accelerating mixed-precision training and inference on massive transformer and HPC models.
Accelerating mixed-precision training and inference on massive transformer and HPC models.
1.98
NVLink Bandwidth
900
GB/s
Linking multiple H100 SXM GPUs via NVLink Switch for low-latency, high-bandwidth multi-GPU scaling.
NVLink Bandwidth
900
Linking multiple H100 SXM GPUs via NVLink Switch for low-latency, high-bandwidth multi-GPU scaling.
Linking multiple H100 SXM GPUs via NVLink Switch for low-latency, high-bandwidth multi-GPU scaling.
900
"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
AI Apps
"Runpod is the only place I can deploy high-end GPU models instantly—no sales calls, no rate limits, no nonsense."
Daniel Chang
Research
“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
LLM Inference
“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 Shimura
Model Training
Comparison
Powerful GPUs. Globally available. Reliability you can trust.
30+ GPUs, 31 regions, instant scale. Fine-tune or go full Skynet—we’ve got you.
Community Cloud
$
2.69
/hr
N/A
Secure Cloud
$
2.99
/hr
Unique GPU Models
25
19
Unique GPU Models
Lorem ipsum
19
Lorem ipsum
25
Global Regions
17
14
Global Regions
Lorem ipsum
14
Lorem ipsum
17
Network Storage
Network Storage
Lorem ipsum
Lorem ipsum
Lorem ipsum
Lorem ipsum
Enterprise-Grade Reliability
Network Storage
Lorem ipsum
Lorem ipsum
Lorem ipsum
Lorem ipsum
Savings Plans
Network Storage
Lorem ipsum
Lorem ipsum
Lorem ipsum
Lorem ipsum
24/7 Support
Network Storage
Lorem ipsum
Lorem ipsum
Lorem ipsum
Lorem ipsum
Delightful Dev Experience
Network Storage
Lorem ipsum
Lorem ipsum
Lorem ipsum
Lorem ipsum
