Compiler-Grounded Hierarchical Diagnosis for LLM-Based Triton Kernel Optimization

Recent advances in large language models (LLMs) have enabled automated kernel generation and optimization, but most existing approaches rely on surface signals such as compilation feedback and prof…

hgpu.org

Harness Engineering for LLM-Driven GPU Kernel Generation

Large language models (LLMs) can assist GPU kernel generation, but their practical effectiveness depends on whether generated code can be reliably constrained, validated, profiled, and selected. Th…

hgpu.org

PortLBM: A Portable Lattice Boltzmann Tool Leveraging SYCL on AMD, NVIDIA, and Intel GPUs

The lattice Boltzmann method (LBM) is a well-established approach for simulating fluid flows at the mesoscopic scale. With the stagnation of Moore’s law, high-performance computing has shifte…

hgpu.org

FlashPDE: A Drop-In Fused Triton Operator Library for Neural PDE Solvers

Physics-Informed Neural Networks (PINNs) solve PDEs by incorporating physical constraints into neural-network training, but large-scale problems are limited by automatic-differentiation memory over…

hgpu.org

NVIDIA-labs OO Agents: Native Python Object-Oriented Agents

Traditional agent development is split across prompt templates, tool schemas, callback code, and workflow graphs. We present NVIDIA Object-Oriented Agents (NOOA), a model-agnostic Python framework …

hgpu.org

Enhancing the Performance Analysis of NCCL GPU Collectives

Efficient inter-GPU communication is very important for scalable distributed deep learning, yet the internal behaviour of NVIDIA’s Collective Communication Library (NCCL) at the GPU kernel level re…

hgpu.org

Real FP4 Tensor-Core Code in Pure Rust on a Gaming GPU – with NVIDIA’s Own Compiler

We report a viability result: an entire Llama-class decoder, written in pure Rust and compiled to PTX by NVIDIA’s own experimental first-party Rust→PTX backend (cuda-oxide), runs FP4-quantize…

hgpu.org

UniCoder: Unified Visual-to-Code Generation via Symbolic Rewards and Reference-Guided Code Optimization

Visual-to-Code generation, which transforms scientific plots, vector graphics, and webpages into executable scripts, demands a level of pixel-precise alignment that standard Multimodal Large Langua…

hgpu.org

Augmenting LLM Code Translation with Compiler Analysis for C to Triton Kernel Generation

Emerging programming models like Triton enable developers to better exploit modern accelerators, but translating legacy code to Triton remains challenging. While Large Language Models (LLMs) show p…

hgpu.org

CuFuzz: An API-Knowledge-Graph Coverage-Driven Fuzzing Framework for CUDA Libraries

In the AI-driven era, NVIDIA CUDA libraries have become indispensable for accelerating compute-intensive tasks, yet their security assessment remains critically understudied due to closed-source co…

hgpu.org

Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization

We present KernelPro, a closed-loop multi-agent system that automatically generates, profiles, and iteratively optimizes GPU kernel code by integrating large language model (LLM) code generation wi…

hgpu.org

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation

Agentic kernel optimization automates manual GPU kernel tuning via iterative generation, validation, and profiling with reasoning LLMs, casting the optimization task as feedback-guided search. Howe…

hgpu.org

The Correctness Illusion in LLM-Generated GPU Kernels

Benchmarks for LLM-generated GPU kernels (KernelBench, TritonBench, GEAK) score correctness through fixed-shape, small-sample allclose-style checks. The number of inputs varies between benchmarks. …

hgpu.org

Probe-and-Refine Tuning of Repository Guidance for Coding Agents

LLM-based coding agents need higher-level operational knowledge about a repository (which files house which subsystems, how to run the test suite, which workflows have historically led to wrong fix…

hgpu.org

AutoPass: Evidence-Guided LLM Agents for Compiler Performance Tuning

Large Language Models (LLMs) show promise for code compilation tasks, but applying them to runtime performance tuning is difficult due to complex microarchitectural effects and noisy runtime measur…

hgpu.org

daVinci-kernel: Co-Evolving Skill Selection, Summarization, and Utilization via RL for GPU Kernel Optimization

GPU kernel optimization represents a paradigm where functional correctness is assumed and execution efficiency is the objective. We present daVinci-kernel, a reinforcement learning framework that c…

hgpu.org

Fearless Concurrency on the GPU

Rust has made safe systems programming practical on the CPU, but writing custom GPU kernels in Rust still forces programmers outside the language’s ownership guarantees. We present cuTile Rus…

hgpu.org

Tangram: Hiding GPU Heterogeneity for Efficient LLM Parallelization

The scale of LLM training jobs requires parallelization planning over large GPU clusters. Due to different GPU types and interconnects added over time, these GPU clusters are increasingly heterogen…

hgpu.org

From Tokens to Regions: CUDA-Sensitive Instruction Tuning for GPU Kernel Generation

High-performance CUDA kernels are essential for scalable AI systems, while Large Language Models (LLMs) still struggle to generate correct kernels due to strict and implicit execution constraints. …

hgpu.org

Leveraging AI Ecosystem for Portable and Sustainable GPU Kernels in HPC

High-Performance Computing (HPC) applications increasingly depend on GPUs, yet developing optimized kernels across evolving GPU architectures remains a major productivity bottleneck. With a tile-ba…

hgpu.org

Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation

Large language models (LLMs) have shown strong empirical gains as self-evolving agents for CUDA kernel generation, driven by feedback-conditioned planning across generations. However, how planning …

hgpu.org

CodegenBench: Can LLMs Write Efficient Code Across Architectures?

While large language models (LLMs) have been extensively evaluated on code generation tasks for general-purpose programming and GPU-accelerated environments (e.g., PyTorch, CUDA), their capabilitie…

hgpu.org

KForge: LLM-Driven Cross-Platform Kernel Generation for AI Accelerators

Production inference increasingly targets a heterogeneous mix of accelerators. Agentic pipelines interleave reasoning, tool calls, and multi-agent coordination, each with distinct compute and memor…

hgpu.org

MusaCoder: Native GPU Kernel Generation with Full-Stack Training on Moore Threads GPU

Native GPU kernel generation turns high-level tensor programs into executable, efficient low-level code. Existing Large Language Models (LLMs) struggle with this task, while execution-based reinfor…

hgpu.org

Autonomous heterogeneous catalyst discovery with a self-evolving multi-agent digital twin

Theoretical heterogeneous catalysis promises rapid catalyst discovery, yet computational and machine-learning predictions often deviate from experiment and stay confined to narrow material families…

hgpu.org

Pretraining large language models with MXFP4 on Native FP4 Hardware

Why does full-pipeline FP4 training of large language models often diverge, even when forward activations and activation gradients remain stable? We address this question through a controlled study…

hgpu.org

Analyzing the Impact of Kernel Fusion on GPU Tensor Operation Performance: A Systematic Performance Study

Large numbers of small tensor kernels are executed by GPUs in modern deep learning frameworks, where total performance is frequently constrained by memory bandwidth and kernel launch overheads. Sys…

hgpu.org

CUDABeaver: Benchmarking LLM-Based Automated CUDA Debugging

Debugging CUDA programs has long been challenging because failures often arise from subtle interactions among hardware behavior, compiler decisions, memory hierarchy, and asynchronous execution. Mo…

hgpu.org

Source-to-Source Transformations for GPU Code Generation

GPUs have become essential in modern high performance computing, but programming them correctly remains a significant challenge. This difficulty arises from subtle concurrency bugs that result from…

hgpu.org

KernelBenchX: A Comprehensive Benchmark for Evaluating LLM-Generated GPU Kernels

LLM-based Triton kernel generation has attracted significant interest, yet a fundamental empirical question remains unanswered: where does this capability break down, and why? We present KernelBenc…

hgpu.org