Vast.ai Our sponsor
USA 6 configs 1x-2x
Professional workstation GPU for AI development and CAD.
You can start using a RTX A6000 for $0.35/hr per GPU (on-demand). It's a low-risk price point for smaller workloads compared to the market high of $2.12/hr. Spot instances start lower, at $0.20/hr per GPU.
USA 6 configs 1x-2x
USA 19 configs 1x-8x
USA 1 config 1x
USA 1 config 1x
USA 5 configs 1x-8x
France 5 configs 1x-4x
France 1 config 1x
UAE 4 configs 1x-8x
Singapore 2 configs 1x
USA 1 config 1x
UK 3 configs 1x
UK 1 config 1x
USA 1 config 1x
USA 12 configs 1x-4x
USA 4 configs 8x
Heads up: We do our best to keep these specs & prices accurate. However, cloud costs may fluctuate based on region, usage, and other factors not listed here. These are estimates based on common setups and are for informational purposes only. Always verify current rates & exact specs with the provider before provisioning.
48GB GDDR6 with NVLink (112 GB/s). Professional-grade with ECC memory and ISV certifications. Strong for both AI development and CAD/visualization workloads.
Priced as a professional workstation GPU. For pure AI inference, the L40S offers similar VRAM with better FP8 performance at lower cost.
The median on-demand price has held steady around $0.56/hr per GPU across providers.
With 48GB of VRAM, the RTX A6000 can typically run models up to about 30B parameters in FP16, or 70B-class models in 4-bit quantized form for inference.
The RTX A6000 has 48GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.
The RTX A6000 has 768 GB/s of memory bandwidth. Higher bandwidth helps with faster data transfer between GPU memory and compute cores.
The RTX A6000 supports 6 precision formats. Training: BF16, FP16, TF32, FP32. Inference: INT4, INT8.
Yes. The RTX A6000 supports NVLink with 112 GB/s of bidirectional bandwidth. This helps accelerate multi-GPU communication.
RTX A6000 pricing currently ranges from $0.20/hr to $2.12/hr per GPU, depending on the provider, instance type, and billing model.
At 720 hours per month, one RTX A6000 can cost between $145.94 to $1,528.56 per month, depending on the provider. Reserved and spot pricing can lower that further.
The RTX A6000 is available from 18 cloud providers, including Verda, Runpod, Sesterce. Pricing and availability vary by region and billing model.
Yes. We currently track 100 RTX A6000 listings across 18 cloud providers:
| Billing type | Listings | Avg $/GPU/hr |
|---|---|---|
| On-demand | 46 | $0.71/hr |
| Reserved | 43 | $0.62/hr |
| Spot | 11 | $0.39/hr |
| GPU Memory | 48 GB GDDR6 |
| Memory Interface | 384-bit |
| Memory Bandwidth | 768 GB/s |
| Error-Correcting Code (ECC) | Yes |
| NVIDIA Ampere Architecture-based CUDA Cores | 10,752 |
| NVIDIA Third-Generation Tensor Cores | 336 |
| NVIDIA Second-Generation RT Cores | 84 |
| Single-Precision Performance | 38.7 TFLOPS |
| RT Core Performance | 75.6 TFLOPS |
| Tensor Performance | 309.7 TFLOPS |
| NVIDIA NVLink | Connects two NVIDIA RTX A6000 GPUs |
| NVIDIA NVLink Bandwidth | 112.5 GB/s (bidirectional) |
| System Interface | PCI Express 4.0 x16 |
| Power Consumption | Total board power: 300 W |
| Thermal Solution | Active |
| Form Factor | 4.4” H x 10.5” L, dual slot, full height |
| Display Connectors | 4x DisplayPort 1.4a |
| Max Simultaneous Displays | 4x 4096 x 2160 @ 120 Hz, 4x 5120 x 2880 @ 60 Hz, 2x 7680 x 4320 @ 60 Hz |
| Power Connector | 1x 8-pin CPU |
| Encode/Decode Engines | 1x encode, 2x decode (+AV1 decode) |
| VR Ready | Yes |
| vGPU Software Support | NVIDIA vPC/vApps, NVIDIA RTX Virtual Workstation, NVIDIA Virtual Compute Server |
| vGPU Profiles Supported | 1 GB, 2 GB, 3 GB, 4 GB, 6 GB, 8 GB, 12 GB, 16 GB, 24 GB, 48 GB |
| Graphics APIs | DirectX 12.07, Shader Model 5.17, OpenGL 4.68, Vulkan 1.18 |
| Compute APIs | CUDA, DirectCompute, OpenCL™ |
Source: official Nvidia RTX A6000 datasheet.
Last updated