NVIDIA RTX A6000

NVIDIA48GB GDDR6PCIe Gen4Ampere

Data updated 2026-11-04 21:38 UTC · refreshed every 5 min

Current Price

$0.69/hr

From $0.27/hr · lowest live offer

23 GPU · Good availability

Current Price

$0.69

-2.8% (24h)

Available Now

23 GPU

Good availability

Total in Network

169 GPU

Live data

Utilization

86.4%

High

Market history

Price Tracking

USD/hour (minimum · median)

Available GPUs

Ready to rent

Total in Network

Infrastructure Growth

Which LLMs run on one NVIDIA RTX A6000

Open VRAM calculator with RTX A6000

Estimated with our VRAM calculator for a single card: 8K context, 4 concurrent users, vLLM-style serving.

Largest that fitsINT4 / AWQ

Qwen3.6 35B A3B

35B MoE · 3B active

23.8 of 48 GB49%

~70 tok/s per stream

Details

Where the memory goes

  • Model weights21.88 GB
  • KV cache0.67 GB
  • Activations0.01 GB
  • CUDA & framework1.19 GB
Precision
INT4 / AWQ
Serving preset
8K context × 4 users
Users at this context
148
Spare context in leftover VRAM
982K
Fit score
85/100 · Moderate
Open in VRAM calculator
BalancedFP16

Qwen3-14B

14B

35.6 of 48 GB74%

~20 tok/s per stream

Details

Where the memory goes

  • Model weights28.00 GB
  • KV cache6.44 GB
  • Activations0.13 GB
  • CUDA & framework1.07 GB
Precision
FP16
Serving preset
8K context × 4 users
Users at this context
11
Spare context in leftover VRAM
71K
Fit score
68/100 · High
Open in VRAM calculator
CompactFP16

Qwen3.5-4B

4B

11.4 of 48 GB24%

~59 tok/s per stream

Details

Where the memory goes

  • Model weights9.20 GB
  • KV cache1.07 GB
  • Activations0.06 GB
  • CUDA & framework1.02 GB
Precision
FP16
Serving preset
8K context × 4 users
Users at this context
140
Spare context in leftover VRAM
1.0M
Fit score
95/100 · Comfortable
Open in VRAM calculator

Estimates, not benchmarks: real usage depends on the runtime, context and batch settings.

Technical Specifications

Memory

48GB GDDR6

FP32 Performance

38.7 TFLOPS

Tensor Performance

155 TFLOPS

Memory Bandwidth

768 GB/s

Architecture

Ampere

TDP

300W

FP64 Performance

0.6 TFLOPS

CUDA Cores

10752

Tensor Cores

336

Ideal Use Cases

AI TrainingDeep LearningProfessional VisualizationCAD/CAMScientific Computing

About NVIDIA RTX A6000

The RTX A6000 is a professional-grade GPU with 48GB memory for demanding AI workloads.

The NVIDIA RTX A6000 features 48GB of GDDR6 memory with 768 GB/s bandwidth, making it excellent for ai training, deep learning, professional visualization. With 155 TFLOPS tensor performance and 38.7 TFLOPS FP32 compute, this NVIDIA Ampere GPU delivers exceptional performance for demanding workloads.

Currently, 169 GPU RTX A6000 GPUs are available in the GPUniq network with 86.4% utilization. The average rental price is $0.69 per hour, with 23 GPU ready for immediate deployment.

RTX A6000 vs similar GPUs — specs & live rental prices

Prices are live minimum / median USD per hour on the GPUniq marketplace, as of 2026-11-04 21:38 UTC.

GPUVRAMArchitectureFP32TensorBandwidthTDPMin $/hrMedian $/hrAvailable
RTX A600048GB GDDR6Ampere38.7 TFLOPS155 TFLOPS768 GB/s300W$0.35$0.6920
RTX 6000 Ada48GB GDDR6Ada Lovelace91.1 TFLOPS1457 TFLOPS960 GB/s300W$0.90$0.9725
RTX 5880 Ada48GB GDDR6Ada Lovelace69.3 TFLOPS1108 TFLOPS960 GB/s285W$0.74$0.8018
RTX A500024GB GDDR6Ampere27.8 TFLOPS111 TFLOPS768 GB/s230W$0.28$0.3530

Frequently Asked Questions

How much does it cost to rent NVIDIA RTX A6000?

The current rental price for NVIDIA RTX A6000 is $0.69 per hour on GPUniq. Prices may vary based on demand and availability.

What is NVIDIA RTX A6000 best used for?

NVIDIA RTX A6000 is ideal for AI Training, Deep Learning, Professional Visualization, CAD/CAM, Scientific Computing. With 48GB of GDDR6 memory and 155 TFLOPS tensor performance, it excels at demanding AI and compute workloads.

How many NVIDIA RTX A6000 GPUs are available for rent?

Currently, there are 23 GPU available for immediate rental out of 169 GPU total in the GPUniq network. The current utilization is 86.4%.

Which LLMs can run on NVIDIA RTX A6000?

On a single NVIDIA RTX A6000, our VRAM calculator fits Qwen3.6 35B A3B (INT4 / AWQ, 23.8 of 48 GB); Qwen3-14B (FP16, 35.6 of 48 GB); Qwen3.5-4B (FP16, 11.4 of 48 GB) — estimated at 8K context with 4 concurrent users.

What are the specifications of NVIDIA RTX A6000?

NVIDIA RTX A6000 features 48GB GDDR6 memory, 38.7 TFLOPS FP32 performance, 155 TFLOPS tensor performance, 768 GB/s memory bandwidth, and 300W TDP. It uses the Ampere architecture.

Related GPUs in RTX Professional Series

GPU rental guides

NVIDIA official specificationsAPI docsAbout GPUniqContact supportPrivacy policyTerms of use

$0.69/hr

RTX A6000 · 23 GPU

Rent now