AscendCraft: Automatic Ascend NPU Kernel Generation via DSL-Guided Transcompilation

• URL: http://arxiv.org/abs/2601.22760v1
• Date: Fri, 30 Jan 2026 09:34:59 GMT
• Title: AscendCraft: Automatic Ascend NPU Kernel Generation via DSL-Guided Transcompilation
• Authors: Zhongzhen Wen, Shudi Shao, Zhong Li, Yu Ge, Tongtong Xu, Yuanyi Lin, Tian Zhang,
• Abstract summary: We present AscendCraft, a DSL-guided approach for automatic AscendC kernel generation.<n>AscendingCraft achieves 98.1% compilation success and 90.4% functional correctness.<n>We also demonstrate that DSL-guided transcompilation can enable LLMs to generate both correct and competitive NPU kernels.
• Score: 8.878393510726008
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The performance of deep learning models critically depends on efficient kernel implementations, yet developing high-performance kernels for specialized accelerators remains time-consuming and expertise-intensive. While recent work demonstrates that large language models (LLMs) can generate correct and performant GPU kernels, kernel generation for neural processing units (NPUs) remains largely underexplored due to domain-specific programming models, limited public examples, and sparse documentation. Consequently, directly generating AscendC kernels with LLMs yields extremely low correctness, highlighting a substantial gap between GPU and NPU kernel generation. We present AscendCraft, a DSL-guided approach for automatic AscendC kernel generation. AscendCraft introduces a lightweight DSL that abstracts non-essential complexity while explicitly modeling Ascend-specific execution semantics. Kernels are first generated in the DSL using category-specific expert examples and then transcompiled into AscendC through structured, constraint-driven LLM lowering passes. Evaluated on MultiKernelBench across seven operator categories, AscendCraft achieves 98.1% compilation success and 90.4% functional correctness. Moreover, 46.2% of generated kernels match or exceed PyTorch eager execution performance, demonstrating that DSL-guided transcompilation can enable LLMs to generate both correct and competitive NPU kernels. Beyond benchmarks, AscendCraft further demonstrates its generality by successfully generating two correct kernels for newly proposed mHC architecture, achieving performance that substantially surpasses PyTorch eager execution.

Related papers

• CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation [51.72529978689561]
  Agent is a large-scale agentic reinforcement learning system that develops kernel expertise through three components.<n>Agent delivers 100%, 100%, and 92% faster rate over torchcompile on KernelBench.
  arXiv  Detail & Related papers  (2026-02-27T18:58:05Z)
• AscendKernelGen: A Systematic Study of LLM-Based Kernel Generation for Neural Processing Units [39.846358001824996]
  We propose Ascend KernelGen, a generation-evaluation integrated framework for NPU kernel development.<n>We introduce Ascend-CoT, a high-quality dataset incorporating chain-of-thought reasoning derived from real-world kernel implementations.<n>We also design NPU KernelBench, a comprehensive benchmark for assessing compilation, correctness, and performance across varying complexity levels.
  arXiv  Detail & Related papers  (2026-01-12T03:12:58Z)
• Library Liberation: Competitive Performance Matmul Through Compiler-composed Nanokernels [37.00431889602245]
  This paper introduces a compilation scheme that automatically generates scalable, high-performance micro Kernels.<n>We implement this technique in an MLIR-based compiler supporting both vector and tile based CPU instructions.<n>Experiments show that the generated nano Kernels are of production-quality, and competitive with state-of-the-art micro Kernel libraries.
  arXiv  Detail & Related papers  (2025-11-14T14:32:28Z)
• Astra: A Multi-Agent System for GPU Kernel Performance Optimization [10.715861478214961]
  We introduce Astra, the first multi-agent system for GPU kernel optimization.<n>Within Astra, specialized agents collaborate through code generation, profiling, and planning to produce kernels that are both correct and high-performance.
  arXiv  Detail & Related papers  (2025-09-09T08:39:50Z)
• KernelBench: Can LLMs Write Efficient GPU Kernels? [36.4117525096377]
  KernelBench is an open-source framework for evaluating language models' ability to write fast and correct kernels.<n>We introduce a new evaluation metric fast_p, which measures the percentage of generated kernels that are functionally correct.<n>Our experiments show that frontier reasoning models perform the best out of the box but still fall short overall.
  arXiv  Detail & Related papers  (2025-02-14T19:30:53Z)
• EPS-MoE: Expert Pipeline Scheduler for Cost-Efficient MoE Inference [49.94169109038806]
  This paper introduces EPS-MoE, a novel expert pipeline scheduler for MoE that surpasses the existing parallelism schemes.<n>Our results demonstrate at most 52.4% improvement in prefill throughput compared to existing parallel inference methods.
  arXiv  Detail & Related papers  (2024-10-16T05:17:49Z)
• Harnessing Deep Learning and HPC Kernels via High-Level Loop and Tensor Abstractions on CPU Architectures [67.47328776279204]
  This work introduces a framework to develop efficient, portable Deep Learning and High Performance Computing kernels. We decompose the kernel development in two steps: 1) Expressing the computational core using Processing Primitives (TPPs) and 2) Expressing the logical loops around TPPs in a high-level, declarative fashion. We demonstrate the efficacy of our approach using standalone kernels and end-to-end workloads that outperform state-of-the-art implementations on diverse CPU platforms.
  arXiv  Detail & Related papers  (2023-04-25T05:04:44Z)
• Kernel Identification Through Transformers [54.3795894579111]
  Kernel selection plays a central role in determining the performance of Gaussian Process (GP) models. This work addresses the challenge of constructing custom kernel functions for high-dimensional GP regression models. We introduce a novel approach named KITT: Kernel Identification Through Transformers.
  arXiv  Detail & Related papers  (2021-06-15T14:32:38Z)
• PolyDL: Polyhedral Optimizations for Creation of High Performance DL primitives [55.79741270235602]
  We present compiler algorithms to automatically generate high performance implementations of Deep Learning primitives. We develop novel data reuse analysis algorithms using the polyhedral model. We also show that such a hybrid compiler plus a minimal library-use approach results in state-of-the-art performance.
  arXiv  Detail & Related papers  (2020-06-02T06:44:09Z)
• PolyScientist: Automatic Loop Transformations Combined with Microkernels for Optimization of Deep Learning Primitives [55.79741270235602]
  We develop a hybrid solution to the development of deep learning kernels. We use the advanced polyhedral technology to automatically tune the outer loops for performance.
  arXiv  Detail & Related papers  (2020-02-06T08:02:34Z)
• PatDNN: Achieving Real-Time DNN Execution on Mobile Devices with Pattern-based Weight Pruning [57.20262984116752]
  We introduce a new dimension, fine-grained pruning patterns inside the coarse-grained structures, revealing a previously unknown point in design space. With the higher accuracy enabled by fine-grained pruning patterns, the unique insight is to use the compiler to re-gain and guarantee high hardware efficiency.
  arXiv  Detail & Related papers  (2020-01-01T04:52:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.