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

arXiv: https://arxiv.org/abs/2601.22760 · HTML: https://mlsanity.com/paper/2601.22760/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 19 references that are in the map.
- First posted on arXiv 30 Jan 2026.

## Builds on (most cited first, 12 of 12)

- [DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence](https://mlsanity.com/paper/2406.11931/) (2024)
- [KernelBench: Can LLMs Write Efficient GPU Kernels?](https://mlsanity.com/paper/2502.10517/) (2025)
- [Kevin: Multi-Turn RL for Generating CUDA Kernels](https://mlsanity.com/paper/2507.11948/) (2025)
- [Astra: A Multi-Agent System for GPU Kernel Performance Optimization](https://mlsanity.com/paper/2509.07506/) (2025)
- [AutoTriton: Automatic Triton Programming with Reinforcement Learning in LLMs](https://mlsanity.com/paper/2507.05687/) (2025)
- [MultiKernelBench: A Multi-Platform Benchmark for Kernel Generation](https://mlsanity.com/paper/2507.17773/) (2025)
- [TritonRL: Training LLMs to Think and Code Triton Without Cheating](https://mlsanity.com/paper/2510.17891/) (2025)
- [QiMeng-Kernel: Macro-Thinking Micro-Coding Paradigm for LLM-Based High-Performance GPU Kernel Generation](https://mlsanity.com/paper/2511.20100/) (2025)
- [EvoEngineer: Mastering Automated CUDA Kernel Code Evolution with Large Language Models](https://mlsanity.com/paper/2510.03760/) (2025)
- [PRAGMA: A Profiling-Reasoned Multi-Agent Framework for Automatic Kernel Optimization](https://mlsanity.com/paper/2511.06345/) (2025)
- [AKG kernel Agent: A Multi-Agent Framework for Cross-Platform Kernel Synthesis](https://mlsanity.com/paper/2512.23424/) (2025)
- [Sharpen the Spec, Cut the Code: A Case for Generative File System with SYSSPEC](https://mlsanity.com/paper/2512.13047/) (2025)

## Built on by (most cited first, 7 of 7)

- [AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization](https://mlsanity.com/paper/2603.23566/) (2026)
- [Hawk: Harnessing Hardware-Aware Knowledge for High-Performance NPU Kernel Generation](https://mlsanity.com/paper/2607.01590/) (2026)
- [SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD](https://mlsanity.com/paper/2607.20145/) (2026)
- [CANN Bench: Benchmarking Agent Generated Kernels against Real NPU and Algorithmic Limits](https://mlsanity.com/paper/2607.20518/) (2026)
- [Compiler-Grounded Hierarchical Diagnosis for LLM-Based Triton Kernel Optimization](https://mlsanity.com/paper/2607.23089/) (2026)
- [Don't Regenerate, Debug: A Domain-Specific Agent for Repairing Near-Miss Hardware Operators](https://mlsanity.com/paper/2608.02712/) (2026)
- [From Experience to Expertise: Adoption-Aware Memory Learning for Data-Scarce NPU Kernel Synthesis](https://mlsanity.com/paper/2609.35568/) (2026)

## Abstract (arXiv)

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.