| Product | NVIDIA® T400 4GB GDDR6 - PCIe 3.0 x16 - Active Cooling (3x mDP) | NVIDIA® RTX A6000 - 48GB GDDR6 - PCIe 4.0 x16 - Active Cooling (4xDP) | NVIDIA® RTX 5000 Ada Generation - 32GB GDDR6 ECC - PCIe 4.0 x16 - Active Cooling (4xDP) |
| Action | Select | Select | Select |
| Main Specifications | |||
| Host Interface | PCI Express 3.0 x16 | PCI Express 4.0 x16 | PCI Express 4.0 x16 |
| Product Type | Workstation | Workstation | Workstation |
| Product Line | NVIDIA Professional Graphics | NVIDIA Professional Graphics | |
| Memory Technology | GDDR6 | GDDR6 | GDDR6 |
| Memory Capacity | 4 GB | 48 GB | 32 GB GDDR6 ECC |
| Max Displays | 4 Displays | 4 Displays | |
| Detailed Specifications | |||
| Processor | NVIDIA Turing | Ampere (GA102) | NVIDIA Ada Lovelace |
| Memory Bandwidth | 80 GB/s | 768 GB/s | 576 GB/s |
| Memory Interface | 64-bit | 384-bit | 256-bit |
| Core Clock Speed | 1455 MHz Base Clock 1860 MHz Boost Clock | ||
| Memory Clock Speed | 2000 MHz 16 Gbps effective | ||
| L2 Cache Size | 6 MB | ||
| Peak Single-Precision Performance | 65.3 TFLOPS | ||
| API Support | CUDA C, CUDA C++, DirectCompute 5.0, OpenCL, Java, Python, and Fortran Shader Model 5.1 (OpenGL 4.5 and DirectX 12) | CUDA 8.5, OpenCL 2.0 Shader Model 6.5, OpenGL 4.6, DirectX 12 Ultimate (12_2), Vulkan 1.2 | |
| Texture Fill Rate | 625 GTexel/s | ||
| Transistor Count | 28.3 Billion | 76.3 billion | |
| Graphics Resolution | Max Digital Resolution: 7680 x 4320 at 60 Hz | 7680 x 4320 x36 bpp at 60 Hz | |
| Streaming Processor Cores | 384 CUDA Parallel-Processing Cores | 10752 Shading Units | 12,800 CUDA Parallel Processing Cores |
| NVIDIA Tensor Cores | 336 | 400 | |
| Tensor Performance | 1044.4 TFLOPS | ||
| Peak Double Precision FP64 Performance | 1,250 GFLOPS (1:32) | ||
| Peak Single Precision FP32 Performance | 1.094 TFLOPS | 38.7 TFLOPS | |
| Peak Half Precision FP16 Performance | 40.00 TFLOPS (1:1) | ||
| Multi-GPU Scalability | NVLINK 2-way low profile (2-slot and 3-slot bridges) connects 2x NVIDIA RTX A6000 | ||
| NVLink Interconnect | 112.5 GB/s (bidirectional) | Not Supported | |
| NVIDIA RT Cores | 84 | 100 | |
| RT Core Performance | 151.0 TFLOPS | ||
| VR Ready | Yes | ||
| Vulkan API | 1.2 | ||
| NVIDIA CUDA™ Technology | Yes | ||
| DisplayPort Output | 4x DisplayPort 1.4a | ||
| Mini DisplayPort Output | x3 | 4x DP 1.4a | |
| Max Graphics Card Power (W) | 30W | 300W | 250W |
| Minimum Recommended Power, Single Card (W) | 700W | ||
| Minimum Recommended Power, 2-Way (W) | 850 | ||
| Minimum Recommended Power, 3-Way (W) | 1000 | ||
| Minimum Recommended Power, 4-Way (W) | 1200 | ||
| Supplementary Power Connectors | 1x 8-pin EPS | 1x 16-pin CEM5 PCIe | |
| Thermal Solution | Ultra-quiet active fansink | Active Heatsink | Blower Active Fan |
| Slot Height | Low-Profile Single Slot | 2-Slot | 2-Slot |
| Lithography | Samsung 8nm | 4 nm NVIDIA Custom Process | |
| Dimensions | 4.4" (H) x 10.5" (L) | 4.4" H x 10.5" L | |
| Action | Select | Select | Select |