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 Ampere Architecture
Power-efficient workstation architecture delivering up to 2.7X the FP32 performance of previous generation in single-slot form factor.
Third-Generation Tensor Cores
Enhanced AI acceleration with structural sparsity support delivering up to 11X faster training performance for machine learning tasks.
16GB GDDR6 Memory
Substantial memory capacity with 448GB/s bandwidth enables AI model training and inference on medium to large datasets.
Single-Slot Design
Most powerful single-slot professional GPU with 140W power consumption, fitting into compact workstations and space-constrained systems.
Performance
Key specs at a glance.
Performance benchmarks that push AI, ML, and HPC workloads further.
Memory Bandwidth
448
GB/s
FP16 Tensor Performance
19.17
TFLOPS
PCIe Gen4 ×16 Bandwidth
63
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
448
GB/s
Feeding large, high-resolution image and 3D datasets into GPU memory for real-time rendering and AI image-model inference.
Memory Bandwidth
448
GB/s
Feeding large, high-resolution image and 3D datasets into GPU memory for real-time rendering and AI image-model inference.
FP16 Tensor Performance
19.17
TFLOPS
Accelerating convolution and transformer operations in image-generation and classification workloads.
FP16 Tensor Performance
19.17
TFLOPS
Accelerating convolution and transformer operations in image-generation and classification workloads.
PCIe Gen4 ×16 Bandwidth
63
GB/s
High-speed GPU-to-GPU and host-to-device transfers in multi-card image-model training when NVLink isn’t supported.
PCIe Gen4 ×16 Bandwidth
63
GB/s
High-speed GPU-to-GPU and host-to-device transfers in multi-card image-model training when NVLink isn’t supported.
"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
$
0.17
/hr
N/A
Secure Cloud
$
0.25
/hr
Unique GPU Models
25
19
Unique GPU Models
Secure Cloud
19
Community Cloud
25
Global Regions
17
14
Global Regions
Secure Cloud
14
Community Cloud
17
Network Storage
Network Storage
Secure Cloud
✔️
Community Cloud
✖️
Enterprise-Grade Reliability
Enterprise-Grade Reliability
Secure Cloud
✔️
Community Cloud
✖️
Savings Plans
Savings Plans
Secure Cloud
✔️
Community Cloud
✖️
24/7 Support
24/7 Support
Secure Cloud
✔️
Community Cloud
✔️
Delightful Dev Experience
Delightful Dev Experience
Secure Cloud
✔️
Community Cloud
✔️
.svg)