PantheonSim / GPUs / AMD Instinct MI325X
Simulated AMD Instinct MI325X: a GPU profile for CUDA and ROCm testing
PantheonSim, the GPU simulator for CUDA and ROCm, simulates the AMD Instinct MI325X (CDNA 3, gfx942) as the profile amd/mi325x.
The CDNA 3 Instinct accelerator with 256 GiB of HBM. It is the only Instinct profile read from a physical card. Its profile is verified, which means it was read from a physical card.
The profile at a glance
These are the numbers a program can ask this simulated GPU, and the values rocm-smi and amd-smi show for it. They come from the profile file in the simulator's repository, pinned to one commit of it.
| Property | Value |
|---|---|
| Profile id | amd/mi325x |
| Vendor and architecture | AMD, CDNA 3 |
| GPU target | gfx942 |
| Device memory | 255.7 GiB, HBM |
| Multiprocessors (as HIP reports them) | 304 |
| Wavefront size | 64 |
| Max threads per multiprocessor | 2048 |
| Shared memory per block (opt-in maximum) | 64 KiB |
| L2 cache | 4 MiB |
| Power limit (profile) | 1000 W |
| Maximum clocks (profile) | 2100 MHz shader, 0 MHz memory |
| ECC memory | yes |
| PCIe link reported | Gen5 x16 |
| Tensor or matrix cores | no |
| Verified (measured on a physical card) | yes |
What this GPU is for
The CDNA 3 Instinct accelerator with 256 GiB of HBM. It is the only Instinct profile read from a physical card.
CDNA 3 is the architecture of the Instinct MI300 series, built for gfx942. Kernels run in 64-lane wavefronts, and the simulator checks the CDNA 3 code it runs instruction by instruction against ROCm's llvm-objdump.
Among the 10 AMD profiles it ranks 3rd by device memory (255.7 GiB) and 1st by multiprocessors as HIP counts them (304), counting the largest as first.
Is this profile verified?
Yes. This profile is marked verified, which means its values were read from a physical AMD Instinct MI325X by the simulator's characterization tools, not copied from a data sheet. Verified describes the profile's numbers: it says nothing about speed, because the simulator has no timing model.
Simulate this GPU
Build the simulator (see the docs) or open the playground, then:
hipcc --offload-arch=gfx942 my_app.hip -o my_app vgpu run --gpu amd/mi325x ./my_app vgpu shell -y --gpu amd/mi325x --count 4 --rocm 7.1.0 vgpu info --gpu amd/mi325x
Build for this card's own gfx target, gfx942. A program built for another target is refused by name, as the real runtime refuses it, and hipErrorNoBinaryForGpu means exactly this.
To use it every time, put gpu: amd/mi325x in a pantheonsim.yaml settings file, or in a GitHub Actions job give the PantheonSim action gpu: amd/mi325x (see CUDA and HIP tests in GitHub Actions). The simulator checks behaviour and has no timing model, so a run on a physical GPU before a release still counts. The ROCm and HIP guide explains the rest.
Where the numbers come from
Each profile file begins with a note on its sources. This is the note for AMD Instinct MI325X, as written in the file:
Source note for amd/mi325x
Generated by amd/tools/rocminfo-to-profile.py from a physical device. Every value below was read from the hardware, not a datasheet.
The GPU index lists every profile, and this one is measured on a physical card.
Related GPUs
Other CDNA 3 profiles: