nncase: An End-to-End Compiler for Efficient LLM Deployment on Heterogeneous Storage Architectures
This addresses the challenge of high adaptation costs and fragmented workflows in deploying LLMs across diverse hardware, offering an automated solution for improved performance.
The paper tackles the problem of efficient LLM deployment on heterogeneous storage architectures by introducing nncase, an end-to-end compiler that outperforms frameworks like MLC LLM and Intel IPEX on Qwen3 models and matches hand-optimized llama.cpp on CPUs.
The efficient deployment of large language models (LLMs) is hindered by memory architecture heterogeneity, where traditional compilers suffer from fragmented workflows and high adaptation costs. We present nncase, an open-source, end-to-end compilation framework designed to unify optimization across diverse targets. Central to nncase is an e-graph-based term rewriting engine that mitigates the phase ordering problem, enabling global exploration of computation and data movement strategies. The framework integrates three key modules: Auto Vectorize for adapting to heterogeneous computing units, Auto Distribution for searching parallel strategies with cost-aware communication optimization, and Auto Schedule for maximizing on-chip cache locality. Furthermore, a buffer-aware Codegen phase ensures efficient kernel instantiation. Evaluations show that nncase outperforms mainstream frameworks like MLC LLM and Intel IPEX on Qwen3 series models and achieves performance comparable to the hand-optimized llama.cpp on CPUs, demonstrating the viability of automated compilation for high-performance LLM deployment. The source code is available at https://github.com/kendryte/nncase.