Product Compare
AMD Radeon Instinct MI350X vs NVIDIA Tesla K40st
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jun 2025 | Nov 2013 |
| Generation | Radeon Instinct | Tesla |
| Type | Professional | Professional |
| Bus Interface | PCIe 5.0 x16 | PCIe 3.0 x16 |
| Clock Speeds | ||
| Base Clock | 1000 MHz | |
| Boost Clock | 2200 MHz | |
| Memory Clock | 2000 MHz | 1502 MHz |
| Memory | ||
| Memory Size | 288GB | 12GB |
| Memory Type | HBM3e | GDDR5 |
| Memory Bus | 8192bit | 384bit |
| Bandwidth | 8.19TB/s | 288.4GB/s |
| Render Config | ||
| Compute Units | 256 | |
| Shading Units | 16384 | 2880 |
| TMUs | 1024 | 240 |
| ROPs | 0 | 48 |
| Tensor Cores | 1024 | |
| L1 Cache | 16 KB (per CU) | 16 KB (per SMX) |
| L2 Cache | 16 MB | 1536 KB |
| L3 Cache | 256 MB | |
| Theoretical Performance | ||
| Pixel Rate | 0 MPixel/s | 34.50 GPixel/s |
| Texture Rate | 2252.8 GTexel/s | 138.0 GTexel/s |
| FP16 (half) | 72.09 TFLOPS | |
| FP32 (float) | 72.09 TFLOPS | 3.312 TFLOPS |
| FP64 (double) | 36.04 TFLOPS | 1104 GFLOPS |
| Graphics Processor | ||
| GPU Name | MI350 256CU | GK110B |
| Architecture | CDNA 4.0 | Kepler |
| Foundry | TSMC | TSMC |
| Process Size | 3 nm | 28 nm |
| Transistors | 185 billion | 7.08 billion |
| Die Size | Unknown | 561 mm² |
| Board Design | ||
| TDP | 1000W | 245W |
| Suggested PSU | 1400 W | 550 W |
| Outputs | No outputs | No outputs |
| Power Connectors | None | |
| Graphics Features | ||
| DirectX | N/A | 12 (11_1) |
| OpenGL | N/A | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | N/A | 1.1 |
| Shader Model | N/A | 5.1 |
| CUDA | 3.5 | |