kernel-fusion

Burn is a next generation tensor library and Deep Learning Framework that doesn't compromise on flexibility, efficiency and portability.

AMD RAD's multi-GPU Triton-based framework for seamless multi-GPU programming

An agent harness that compiles a model into one provably-correct, self-retargeting CUDA megakernel and self-tunes it past cuBLAS at batch-1 LLM decode, paper: https://arxiv.org/abs/2606.09682

An efficient concurrent graph processing system

GoPTX: Fine-grained GPU Kernel Fusion by PTX-level Instruction Flow Weaving

Fused Triton kernels for TurboQuant KV cache compression — 2-4 bit quantization with RHT rotation. Drop-in HuggingFace & vLLM integration. Up to 4.9x KV cache compression for Llama, Qwen, Mistral, and more.

Noeris — autonomous kernel fusion discovery + Triton autotuning for LLM kernels and Gemma layer deeper fusion (A100/H100 wins).

面向 Transformer 推理学习的精简 Triton 算子：RMSNorm+RoPE、Gated MLP、FlashAttention

LAMB go brrr

MLX + Metal implementation of mHC: Manifold-Constrained Hyper-Connections by DeepSeek-AI.

A CUDA framework for separating graph-recoverable launch overhead from byte-elimination gains in LLM decode-chain fusion

Assigment 3 for the "Parallel & Distributed Systems" course (ECE, AUTh) - Fall 2024

High-performance CUDA implementation of LayerNorm for PyTorch achieving 1.46x speedup through kernel fusion. Optimized for large language models (4K-8K hidden dims) with vectorized memory access, warp-level primitives, and mixed precision support. Drop-in replacement for nn.LayerNorm with 25% memory reduction.

GPU fusion code and algorithm

Write the math. Get the kernel. Fused CUDA kernel generation from mathematical specifications.

Compile time kernels fusion and expression trees as Alpaka boost.odeint backend. This is my team project developed in collaboration with and under the supervision of HZDR.

A torch.compile backend that pays attention: custom FX fusion passes, a Triton LayerNorm kernel, and a persistent compile cache. Apache-2.0 free core of the g2n platform.

Fused STFT + mel filterbank CUDA kernel — ~3x faster log-mel spectrograms than torchaudio's GPU path, verified on NVIDIA GB10 (DGX Spark)

Experiments with optimized inference of Qwen3-ASR with vllm

Production-grade Triton kernel fusing residual add + RMSNorm + packed QKV projection into a single GPU launch for decoder-only transformer inference (Llama-3, Mistral, Qwen2). +2.4% tok/s, -1.5 GB VRAM on A10G.