Ampere GPU
NVIDIA A10G
24GB GDDR6X | 250 TFLOPS FP16 | 600 GB/s bandwidth
From $0.540 /hr
Inference Graphics rendering Video encoding Light training
Specifications
VRAM
24 GB GDDR6X
FP16
250 TFLOPS
FP8
N/A
FP4
N/A
Memory BW
600 GB/s
TDP
150W
Interconnect
PCIe Gen4
Architecture
Ampere
Cloud GPU Pricing (4 offers)
| Provider | Instance Type | vCPUs | RAM | Price/hr | Price/mo | Spot Price | Availability | Action |
|---|---|---|---|---|---|---|---|---|
RunPodCheapest | a10g-24gb | 8 | 62 GB | $0.540/hr | $394.20/mo | $0.380-29.6% | Available | Deploy on RunPod |
Amazon Web Services | g5.4xlarge | 16 | 64 GB | $0.760/hr | $554.80/mo | $0.300-60.5% | Available | Deploy on AWS |
Amazon Web Services | g5.4xlarge | 16 | 64 GB | $0.760/hr | $554.80/mo | $0.270-64.5% | Available | Deploy on AWS |
Amazon Web Services | g5.4xlarge | 16 | 64 GB | $0.840/hr | $613.20/mo | $0.340-59.5% | Available | Deploy on AWS |
Last updated 57h ago
Compare With Other GPUs
Frequently Asked Questions
What is the cheapest NVIDIA A10G cloud provider?
The cheapest NVIDIA A10G is available on RunPod at $0.540/hr (a10g-24gb).
How much does NVIDIA A10G cost per hour?
NVIDIA A10G cloud GPU pricing ranges from $0.540/hr to $0.840/hr depending on the provider and configuration.
What are the specs of NVIDIA A10G?
NVIDIA A10G features 24GB GDDR6X memory, 250 TFLOPS FP16 performance, 600 GB/s memory bandwidth, and 150W TDP. Architecture: Ampere.
Is NVIDIA A10G good for AI training?
Yes, NVIDIA A10G is well-suited for AI training workloads. Key use cases include: Inference, Graphics rendering, Video encoding, Light training.
Related GPUs
A100 SXM4 80GB
80GB HBM2e | Ampere
From $1.10/hr
A100 PCIe 80GB
80GB HBM2e | Ampere
From $1.05/hr
A100 PCIe 40GB
40GB HBM2e | Ampere
From $0.850/hr
L40S
48GB GDDR6X | Ada Lovelace
From $0.820/hr
L4
24GB GDDR6 | Ada Lovelace
From $0.350/hr
T4
16GB GDDR6 | Turing
From $0.220/hr