vixhal-baraiya/qwenGPU

Write a fast kernel and see how you compare against the best humans and AI on gpumode.com

High-performance single-GPU inference for selected model checkpoints and GPUs.

Qwen3.8-27B and Qwen3.6-35B-A3B local inference engine for Vulkan and Metal

Qwen3.8 Local Coding on a 16GB GPU

KernelBench: Can LLMs Write GPU Kernels? - Benchmark + Toolkit with Torch -> CUDA (+ more DSLs)

Qwen3-0.6B megakernel: 527 tok/s decode on RTX 3090 (3.8x faster than PyTorch)

Testbed for LLM inference with cutile-rs.

Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.

A lightweight Python-based GPU architecture simulator that demonstrates how parallel threads, registers, memory, and instructions work on a GPU.