Back to all GPUs

NVIDIA RTX A6000 Cloud GPU Specs

HexGrid Cloud rents the NVIDIA RTX A6000 at $0.55 per GPU hour, billed per minute. The RTX A6000 carries 48GB of GDDR6 memory on the Ampere architecture.

48GB VRAMAmpereDeploys in ~1 minSecure Cloud

HexGrid Cloud price

$0.55/hr

Billed per minute

VRAM

48GB

GDDR6

FP32

38.7 TFLOPS

Compute performance

Market average

Other clouds, list price

Running the RTX A6000 on HexGrid Cloud

What you get when you deploy with us, beyond the hourly rate.

Live in under a minute

Pick a region, hit deploy, and SSH in. No quota requests, no sales call.

Per-minute billing

You pay for the minutes you use. Stop the instance and billing stops with it.

Ready for training day one

CUDA, PyTorch and the usual drivers are preinstalled, or bring your own image.

No egress charges

Move checkpoints and datasets out without a surprise bandwidth bill.

About the NVIDIA RTX A6000

What the RTX A6000 is built for, and where it falls short.

The NVIDIA RTX A6000 is a professional Ampere GPU with 48GB of ECC GDDR6 memory. Its large framebuffer makes it attractive for AI inference, model fine-tuning, rendering, visualization, simulation, and virtual workstation workloads where memory capacity is more important than the latest-generation Tensor performance.

Best suited for

The RTX A6000 combines 48GB of ECC memory with Ampere Tensor Cores and professional visualization features, making it useful for large-memory AI and workstation workloads.

  • LLM inference
  • LLM fine-tuning
  • Image generation
  • 3D rendering
  • CAD and visualization
  • Virtual workstations

Strengths

  • 48GB ECC GPU memory
  • NVLink support
  • Strong professional rendering performance
  • vGPU support
  • Large VRAM capacity for AI workloads

Limitations

  • No FP8 Tensor Core acceleration
  • Older Ampere architecture
  • Lower AI throughput than Ada, Hopper, and Blackwell GPUs

NVIDIA RTX A6000 specifications

Full technical specifications for the NVIDIA RTX A6000.

VendorNVIDIAArchitectureAmpereChipGA102CategoryWorkstation GPUReleased2020Form factorDual-slotInterfacePCIe 4.0 x16

Memory

Memory
48 GB
Type
GDDR6
Bandwidth
768 GB/s
Bus width
384-bit
ECC
Yes

Compute

CUDA cores
10,752
Tensor cores
336
Tensor generation
Gen 3
RT cores
84
Boost clock
1,800 MHz

AI & compute performance

FP32
38.7 TFLOPS
RT
75.6 TFLOPS

Power & physical

TDP
300 W
Power connector
1x 8-pin CPU
Form factor
Dual-slot
Cooling
Active
Length
267 mm

Interconnect

Interface
PCIe 4.0 x16
NVLink
Yes
NVLink bandwidth
112.5 GB/s
Infinity Fabric
No

Virtualization

MIG
No
vGPU
Yes

Media engines

NVENC engines
1
NVDEC engines
2
AV1 encode
No
AV1 decode
Yes

Software

CUDA
Yes
ROCm
No

Precision support

FP64
No
FP32
Yes
TF32
Yes
BF16
Yes
FP16
Yes
FP8
No
INT8
Yes

RTX A6000 frequently asked questions

Common questions about NVIDIA RTX A6000 pricing, specifications and workloads on HexGrid Cloud.

How much VRAM does the NVIDIA RTX A6000 have?

The NVIDIA RTX A6000 has 48GB of GDDR6 memory with ECC and 768 GB/s of memory bandwidth.

Is the RTX A6000 good for AI workloads?

Yes. Its 48GB of VRAM makes it useful for LLM inference, fine-tuning, image generation, and other memory-intensive AI workloads.

Does the RTX A6000 support NVLink?

Yes. Two RTX A6000 GPUs can be connected using NVIDIA NVLink with up to 112.5 GB/s of bidirectional bandwidth.

Does the RTX A6000 support FP8?

No. The RTX A6000 is based on the Ampere architecture and predates fourth-generation Tensor Core FP8 acceleration.

How much does an RTX A6000 cost per hour in the cloud?

RTX A6000 cloud pricing varies by provider, region, and configuration. HexGrid compares available hourly offers on this page.

Other GPUs on HexGrid Cloud

If the RTX A6000 isn't the right fit, these are also available to deploy.

Browse every GPU we rent

Start training on the RTX A6000 today

$0.55 per GPU hour, billed per minute, running in about a minute.

Specification sources