Product Compare
AMD Radeon Instinct MI300A vs NVIDIA H100 PCIe 96 GB
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Dec 2023 | Mar 2023 |
| Generation | Radeon Instinct | Tesla Hopper |
| Type | Desktop | AI GPU |
| Bus Interface | PCIe 5.0 x16 | PCIe 5.0 x16 |
| Clock Speeds | ||
| Base Clock | 1000 MHz | 1665 MHz |
| Boost Clock | 2100 MHz | 1837 MHz |
| Memory Clock | 2525 MHz | 1313 MHz |
| Memory | ||
| Memory Size | 192GB | 96GB |
| Memory Type | HBM3 | HBM3 |
| Memory Bus | 8192bit | 5120bit |
| Bandwidth | 10.3TB/s | 1681GB/s |
| Render Config | ||
| Compute Units | 304 | |
| Shading Units | 19456 | 16896 |
| TMUs | 1216 | 528 |
| ROPs | 0 | 24 |
| Tensor Cores | 1216 | 528 |
| L1 Cache | 16 KB (per CU) | 256 KB (per SM) |
| L2 Cache | 16 MB | 50 MB |
| L3 Cache | 256 MB | |
| SM Count | 132 | |
| Theoretical Performance | ||
| Pixel Rate | 0 MPixel/s | 44.09 GPixel/s |
| Texture Rate | 2554 GTexel/s | 969.9 GTexel/s |
| FP16 (half) | 653.7 TFLOPS | 248.3 TFLOPS |
| FP32 (float) | 81.72 TFLOPS | 62.08 TFLOPS |
| FP64 (double) | 81.72 TFLOPS | 31.04 TFLOPS |
| Graphics Processor | ||
| GPU Name | Aqua Vanjaram | GH100 |
| Architecture | CDNA 3.0 | Hopper |
| Foundry | TSMC | TSMC |
| Process Size | 5 nm | 4 nm |
| Transistors | 153 billion | 80 billion |
| Die Size | 1017 mm² | 814 mm² |
| Board Design | ||
| TDP | 750W | 700W |
| Suggested PSU | 1150 W | 1100 W |
| Outputs | No outputs | No outputs |
| Power Connectors | None | 8-pin EPS |
| Graphics Features | ||
| DirectX | N/A | N/A |
| OpenGL | N/A | N/A |
| OpenCL | 3.0 | 3.0 |
| Vulkan | N/A | N/A |
| Shader Model | N/A | N/A |
| CUDA | 9.0 | |