entry
RTX A2000 GPU
Entry-level professional GPU based on NVIDIA Ampere architecture, designed for compact workstations with support for AI inference and lightweight training workloads.
VRAM 6GB
CUDA Cores 3,328
Tensor Cores 104
TDP 70W
Process 6nm
From
$0.06/hr
across 2 providers
Cloud Pricing
Cheapest on Vast.ai — 24% below avg| Provider | Config | Price / hr | Updated | Source |
|---|---|---|---|---|
2× | $0.06/hr | 4/11/2026 | ||
1× | $0.06/hr | 4/6/2026 | ||
1× | $0.06/hr | 4/18/2026 | ||
1× | $0.12/hr | 4/18/2026 |
Direct from providerVia marketplace
Prices updated daily. Last check: 4/18/2026
Performance
FP32
7.99 TFLOPS
Bandwidth
288 GB/s
Common Use Cases
Light AI inference, CAD/CAM visualization, entry-level ML development
Full Specifications
Hardware
- Manufacturer
- NVIDIA
- Architecture
- Ampere
- CUDA Cores
- 3,328
- Tensor Cores
- 104
- RT Cores
- 26
- Process Node
- 6nm
- TDP
- 70W
Memory & Performance
- VRAM
- 6GB
- Memory Interface
- 192-bit
- Memory Bandwidth
- 288 GB/s
- FP32
- 7.99 TFLOPS
- Release
- 2021