| Product | NVIDIA® A40 GPU Computing Accelerator - 48GB GDDR6 - PCIe 4.0 x16 - Passive Cooling | NVIDIA® L4 ADA GPU Computing Accelerator - 24GB GDDR6X - PCIe 4.0 x16 - Passive Cooling | NVIDIA® Quadro® RTX 6000 - 24GB GDDR6 - PCIe 3.0 x16 - Passive Cooling (4xDP) |
| Action | Select | Select | Select |
| Main Specifications | |||
| Product Series | Nvidia A40 | Nvidia L4 | |
| Core Type | NVIDIA TENSOR | NVIDIA TENSOR | |
| Core Clock Speed | 795 MHz Base | 2040 MHz Boost | ||
| Host Interface | PCI Express 4.0 x16 | PCI Express 4.0 x16 | PCI Express 3.0 x16 |
| GPU Architecture | Ampere | Ada Lovelace | |
| Product Type | Workstation | ||
| Product Line | Quadro | ||
| Memory Technology | GDDR6 | ||
| Memory Capacity | 24 GB | ||
| Max Displays | 4 Displays | ||
| Detailed Specifications | |||
| Streaming Processor Cores | 10752 CUDA Cores | 4608 CUDA Parallel-Processing Cores | |
| NVIDIA Tensor Cores | 336 Tensor Cores | 576 | |
| NVIDIA RT Cores | 84 RT Cores | 72 | |
| Memory Clock Speed | 6251 MHz | ||
| Memory Interface | 384-bit | 192-bit | 384-bit |
| Memory Speeds (GT/s) | 14.5Gbps GDDR6 | ||
| Max Memory Size | 48 GB GDDR6 with error-correcting code (ECC) | 24 GB | |
| Max Memory Bandwidth | 696 GB/s | 300 GB/s | |
| Peak FP32 | 30.3 TFLOPS | ||
| Peak TF32 Tensor Core | 120 TFLOPS | Sparsity | ||
| Peak BFLOAT16 Tensor Core | 242 TFLOPS | Sparsity | ||
| Peak FP16 Tensor Core | 242 TFLOPS | Sparsity | ||
| Peak FP8 Tensor Core | 485 TFLOPS | Sparsity | ||
| Peak INT8 Tensor Core | 485 TOPS | Sparsity | ||
| NVIDIA NVLink™ Interconnect Bandwidth | NVIDIA NVLink 112.5 GB/s (bidirectional) PCIe Gen4 16 GB/s | ||
| NVENC | NVDEC | 2 | 4 | 4 | JPEG Decoders | AV1 Encode and Decode | ||
| Secure Boot with Root of Trust | Yes | ||
| NEBS Ready | Yes | Level 3 | ||
| ECC Protection | On by Default | ||
| DisplayPort Connectors | 3x DisplayPort 1.4 A40 is configured for virtualization by default with physical display connectors disabled. The display outputs can be enabled via management software tools. | None | vGPU Only | |
| Cooling | Passive | Passive | |
| Dual Slot | 2-slot Low-profile | No | |
| Dimensions | 4.4" (H) x 10.5" (L) | 4.40” H x 10.50” L | |
| Form Factor | 6.61” L x 2.71” H (Low-profile) | ||
| Lithography | Samsung 8nm | ||
| Supplementary Power Connectors | 1x 8-pin CPU (EPS12V) | ||
| Max Graphics Card Power (W) | 300W | 72W | 250W |
| Processor | NVIDIA Quadro Turing | ||
| Memory Bandwidth | 624 GB/s | ||
| API Support | CUDA, DirectCompute, OpenCL, OpenACC Shader Model 5.1, OpenGL 4.5, DirectX 12.0, Vulkan 1.0 | ||
| Graphics Resolution | Max Virtual Display Head Resolution: 4096 x 2160 | ||
| RTX-OPS | 80T | ||
| Peak Single Precision FP32 Performance | 14.9 TFLOPS | ||
| Peak Half Precision FP16 Performance | 29.9 TFLOPS | ||
| Peak Half Precision INT8 Performance | 238.9 TOPS | ||
| Deep Learning TFLOPS | 119.4 TFLOPS | ||
| Multi-GPU Scalability | NVLink | ||
| NVLink Interconnect | 100 GB/Sec (bidirectional) | ||
| NVLink GPU Memory | 24 GB GDDR6 ECC (2x RTX 6000 passive) | ||
| Real-Time Ray Tracing | 10 GigaRays/sec | ||
| NVIEW | Yes | ||
| NVIDIA MOSAIC | Yes | ||
| Minimum Recommended Power, Single Card (W) | 500W | ||
| Minimum Recommended Power, 2-Way (W) | 750 | ||
| Minimum Recommended Power, 3-Way (W) | 850 | ||
| Minimum Recommended Power, 4-Way (W) | 1000 | ||
| Thermal Solution | Passive Heatsink | ||
| Slot Height | 2-Slot | ||
| Action | Select | Select | Select |