NPUEval: Optimizing NPU Kernels with LLMs and Open Source Compilers
July 18, 2025 · Declared Dead · 🏛 arXiv.org
"Paper promises code 'coming soon'"
Evidence collected by the PWNC Scanner
Authors
Sarunas Kalade, Graham Schelle
arXiv ID
2507.14403
Category
cs.PL: Programming Languages
Citations
2
Venue
arXiv.org
Last Checked
1 month ago
Abstract
Neural processing units (NPUs) are gaining prominence in power-sensitive devices like client devices, with AI PCs being defined by their inclusion of these specialized processors. Running AI workloads efficiently on these devices requires libraries of optimized kernels. Creating efficient kernels demands expertise in domain-specific C++ with vector intrinsics and in-depth knowledge of the target architecture. Unlike GPU programming, which has had years to mature, NPU programming is new, with smaller and more fragmented developer communities across hardware platforms. This fragmentation poses a challenge when utilizing LLMs to assist in writing NPU kernels, as domain-specific optimized code examples are underrepresented in LLM pre-training data. In this paper we introduce NPUEval -- a benchmark for writing and evaluating NPU kernels, consisting of 102 common operators for machine learning workloads. We evaluate LLM generated code on actual hardware based on both functional correctness and vectorization efficiency using open source compiler tools targeting the AMD NPU. We evaluate a range of state-of-the-art LLMs with a mix of proprietary and open-weight models. Latest reasoning models like DeepSeek R1, show promising results achieving out-of-the-box 50%+ vectorization on select kernels. However, the average score across the entire dataset remains roughly 10% even with compiler feedback and vectorized kernel examples -- showing that this is a challenging dataset even for frontier models. The dataset and evaluation code will be released with a permissive open source license, providing an essential benchmark for advancing research in code generation and NPU kernel optimization.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
📜 Similar Papers
In the same crypt — Programming Languages
R.I.P.
👻
Ghosted
R.I.P.
👻
Ghosted
Tensor Comprehensions: Framework-Agnostic High-Performance Machine Learning Abstractions
R.I.P.
👻
Ghosted
Glow: Graph Lowering Compiler Techniques for Neural Networks
R.I.P.
👻
Ghosted
Learnable Programming: Blocks and Beyond
R.I.P.
👻
Ghosted
Scenic: A Language for Scenario Specification and Scene Generation
R.I.P.
👻
Ghosted
Vandal: A Scalable Security Analysis Framework for Smart Contracts
Died the same way — ⏳ Coming Soon™
R.I.P.
⏳
Coming Soon™
Exploring Simple Siamese Representation Learning
R.I.P.
⏳
Coming Soon™
An Analysis of Scale Invariance in Object Detection - SNIP
R.I.P.
⏳
Coming Soon™
Class-balanced Grouping and Sampling for Point Cloud 3D Object Detection
R.I.P.
⏳
Coming Soon™