PantheonSim: GPU simulator for CUDA and ROCm. Run CUDA and HIP programs unmodified on a CPU, no GPU needed: simulated NVIDIA and AMD GPUs for development, teaching and CI.
Simulate Moore Threads MUSA and NVIDIA CUDA GPUs in Kubernetes — fake MUSA/MTML/CUDA/NVML user-space libs, mthreads-gmi/nvidia-smi tools, and a device-plugin for HAMi-style scheduling without physical GPUs.
Tiny tensor compiler: one tensor program compiled to a GPU simulator (tiny-gpu) and a systolic-array TPU (tiny-tpu), checked bit-exact against numpy.
Landing repository for the GPU Simulator project
PantheonSim, the GPU simulator for CUDA and ROCm, as a GitHub Action: run CUDA and HIP tests on simulated NVIDIA and AMD GPUs on standard runners, no GPU needed.
A deterministic functional and cycle-stepped pipeline-accurate SIMT GPU architecture simulator modeling the synthetic GV-SIM32 ISA.
Twelve CUDA-style kernel puzzles on a simulated GPU. Write a kernel, pick the launch, and see warps, divergence, memory coalescing, bank conflicts, atomics and data races, graded against a par time. Built with Claude Code.
Fork of GPGPU-Sim/Accel-Sim adding a Distributed Shared Memory (thread-block cluster) model ported from ClusterSim, with a timing-validated SM-to-SM interconnect and a measured comparison against conventional shared memory.
Auto-graded exercises in low-level ML systems: real C++, real CUDA-C on a software GPU (no NVIDIA hardware), deterministic measured gates.
To associate your repository with the gpu-simulator topic, visit your repo's landing page and select "manage topics."