| Product | AMD Instinct™ MI350P - 144GB HBM3e - PCIe 5.0 x16 - Passive Cooling - 600W TDP | NVIDIA® H200 NVL GPU - 141GB HBM3e - PCIe 5.0 x16 - Passive Cooling - 600W | AMD Radeon™ AI PRO R9700S - 32GB GDDR6 - PCIe 5.0 x16 - Passive Cooling (4xDP) - 300W | NVIDIA® RTX PRO 6000 Blackwell Server Edition - 96GB GDDR7 - PCIe 5.0 x16 - Passive Cooling (4xDP) - 600W |
| Action | Select | Select | Select | Select |
| Main Specifications | ||||
| Product Series | AMD Instinct MI350 Series | Nvidia H200 | ||
| Core Type | NVIDIA TENSOR | |||
| Core Clock Speed | 2200 MHz | |||
| Host Interface | PCI Express 5.0 x16 | PCI Express 5.0 x16 | PCI Express 5.0 x16 | PCI Express 5.0 x16 |
| GPU Architecture | CDNA4 | Hopper | ||
| Product Type | Workstation | Server | ||
| Product Line | Radeon AI PRO R9000 Series | NVIDIA Professional Graphics | ||
| Memory Technology | GDDR6 with ECC (Linux Only) | GDDR7 | ||
| Memory Capacity | 32 GB | 96 GB with ECC | ||
| Max Displays | 4 Displays | 4 Displays | ||
| Detailed Specifications | ||||
| Streaming Processor Cores | 8,192 | 4096 | 24,064 CUDA Parallel Processing Cores | |
| Matrix Cores | 512 | |||
| Compute Units | 128 | |||
| NVIDIA Tensor Cores | 752 (5th Gen) | |||
| NVIDIA RT Cores | 188 (4th Gen) | |||
| Memory Interface | 256-bit | 512-bit | ||
| Max Memory Size | 144 GB HBM3E | 141 GB | ||
| Max Memory Bandwidth | 4 TB/s | 4.8TB/s | ||
| Peak FP64 | 30 TFLOPS | |||
| Peak FP64 Tensor Core | 60 TFLOPS | |||
| INT8 Tensor Core | 3,341 TFLOPS | |||
| TF32 Tensor Core | 835 TFLOPS | |||
| Peak BFLOAT16 Tensor Core | 1,671 TFLOPS | |||
| Peak FP16 Tensor Core | 1,671 TFLOPS | |||
| Peak FP8 Tensor Core | 3,341 TFLOPS | |||
| Peak Microscaling Four-bit Precision Matrix (MXFP4) Performance | 4.6 PFLOPs | |||
| Peak Microscaling Six-bit Precision Matrix (MXFP6) Performance | 4.6 PFLOPs | |||
| Peak Half Precision (FP16) Performance | 72 TFLOPs | |||
| Peak Half Precision Matrix (FP16) Performance | 1.15 PFLOPs | |||
| Peak Double Precision Matrix (FP64) Performance | 36 TFLOPs | |||
| Peak Double Precision (FP64) Performance | 36 TFLOPs | |||
| Peak INT8 Performance | 2.3 POPs | |||
| Peak bfloat16 | 1.15 PFLOPs | |||
| Transistor Count | 73 Billion | 53.9 Billion | ||
| Cooling | Passive | Passive | ||
| Dual Slot | Double Slot | |||
| Dimensions | 10.5" (267 mm) Board Length | 4.4” H x 10.5” L, FHFL Dual Slot | ||
| Form Factor | PCIe® Add-in Card | Dual-slot air-cooled | ||
| Lithography | TSMC 3nm | 6nm FinFET | 4N NVIDIA Custom Process | ||
| Supplementary Power Connectors | 12V-2x6 | 1x PCIe CEM5 16-pin | ||
| Max Graphics Card Power (W) | 600W TBP (Max); 450W TBP configurable | Up to 600W (configurable) | 300W | 400 W – 600 W (configurable) |
| Processor | AMD RDNA™ 4 | NVIDIA Blackwell Architecture | ||
| Memory Bandwidth | 640 GB/s | Up to 1.6 TB/s | ||
| Core Clock Speed | 2350 MHz (Boost Up to 2920 MHz) | |||
| Graphics Resolution | Up to: 4x 4096 x 2160 (4K DCI) @ 120Hz with DSC 2x 6144 x 3456 (6K) 12-bit HDR @ 60Hz Uncompressed 2x 7680 x 4320 (8K) 12-bit HDR @ 60Hz with DSC 1x 12288 x 6912 (12K) @ 120Hz with DSC | |||
| Peak Single Precision FP32 Performance | 47.8 TFLOPs | |||
| Peak Half Precision FP16 Performance | 47.8 TFLOPs | |||
| DisplayPort Output | 4x DP | |||
| Minimum Recommended Power, Single Card (W) | 750W | 1000 | ||
| Minimum Recommended Power, 2-Way (W) | 1200 | 1600 | ||
| Minimum Recommended Power, 3-Way (W) | 1600 | 2000 | ||
| Minimum Recommended Power, 4-Way (W) | 2000 | 2400 | ||
| Thermal Solution | Passive Cooled | Passive | ||
| Slot Height | 2-Slot | 2-Slot | ||
| Action | Select | Select | Select | Select |