Turing GPU
NVIDIA T4
16GB GDDR6 | 130 TFLOPS FP16 | 320 GB/s bandwidth
From $0.220 /hr
Inference Light AI workloads Video transcoding
Specifications
VRAM
16 GB GDDR6
FP16
130 TFLOPS
FP8
N/A
FP4
N/A
Memory BW
320 GB/s
TDP
70W
Interconnect
PCIe Gen3
Architecture
Turing
Cloud GPU Pricing (5 offers)
| Provider | Instance Type | vCPUs | RAM | Price/hr | Price/mo | Spot Price | Availability | Action |
|---|---|---|---|---|---|---|---|---|
Vast.aiCheapest | t4_16gb | 4 | 16 GB | $0.220/hr | $160.60/mo | $0.160-27.3% | Available | Rent on Vast.ai |
RunPod | t4-16gb | 4 | 16 GB | $0.240/hr | $175.20/mo | $0.170-29.2% | Available | Deploy on RunPod |
Google Cloud Platform | n1-standard-4-t4 | 4 | 15 GB | $0.290/hr | $211.70/mo | $0.120-58.6% | Available | Deploy on Google Cloud |
Amazon Web Services | g4dn.xlarge | 4 | 16 GB | $0.350/hr | $255.50/mo | $0.140-60% | Available | Deploy on AWS |
Microsoft Azure | Standard_NC4as_T4_v3 | 4 | 28 GB | $0.380/hr | $277.40/mo | $0.150-60.5% | Available | Deploy on Azure |
Last updated 57h ago
Compare With Other GPUs
Frequently Asked Questions
What is the cheapest NVIDIA T4 cloud provider?
The cheapest NVIDIA T4 is available on Vast.ai at $0.220/hr (t4_16gb).
How much does NVIDIA T4 cost per hour?
NVIDIA T4 cloud GPU pricing ranges from $0.220/hr to $0.380/hr depending on the provider and configuration.
What are the specs of NVIDIA T4?
NVIDIA T4 features 16GB GDDR6 memory, 130 TFLOPS FP16 performance, 320 GB/s memory bandwidth, and 70W TDP. Architecture: Turing.
Is NVIDIA T4 good for AI training?
NVIDIA T4 is primarily designed for: Inference, Light AI workloads, Video transcoding. For large-scale training, consider higher-tier GPUs.