# Xe-Forge: Multi-Stage LLM-Powered Kernel Optimization for Intel GPU

arXiv: https://arxiv.org/abs/2605.26118 · HTML: https://mlsanity.com/paper/2605.26118/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 42 references that are in the map.
- First posted on arXiv 16 Apr 2026.

## Builds on (most cited first, 19 of 19)

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [KernelBench: Can LLMs Write Efficient GPU Kernels?](https://mlsanity.com/paper/2502.10517/) (2025)
- [Ansor: Generating High-Performance Tensor Programs for Deep Learning](https://mlsanity.com/paper/2006.06762/) (2020)
- [Kevin: Multi-Turn RL for Generating CUDA Kernels](https://mlsanity.com/paper/2507.11948/) (2025)
- [TritonBench: Benchmarking Large Language Model Capabilities for Generating Triton Operators](https://mlsanity.com/paper/2502.14752/) (2025)
- [Astra: A Multi-Agent System for GPU Kernel Performance Optimization](https://mlsanity.com/paper/2509.07506/) (2025)
- [CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation](https://mlsanity.com/paper/2602.24286/) (2026)
- [AutoTriton: Automatic Triton Programming with Reinforcement Learning in LLMs](https://mlsanity.com/paper/2507.05687/) (2025)
- [Geak: Introducing Triton Kernel AI Agent & Evaluation Benchmarks](https://mlsanity.com/paper/2507.23194/) (2025)
- [STARK: Strategic Team of Agents for Refining Kernels](https://mlsanity.com/paper/2510.16996/) (2025)
- [AVO: Agentic Variation Operators for Autonomous Evolutionary Search](https://mlsanity.com/paper/2603.24517/) (2026)
- [SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits](https://mlsanity.com/paper/2603.19173/) (2026)
- [AutoKernel: Autonomous GPU Kernel Optimization via Iterative Agent-Driven Search](https://mlsanity.com/paper/2603.21331/) (2026)
- [TritonForge: Profiling-Guided Framework for Automated Triton Kernel Optimization](https://mlsanity.com/paper/2512.09196/) (2025)
- [KernelFoundry: Hardware-aware evolutionary GPU kernel optimization](https://mlsanity.com/paper/2603.12440/) (2026)
- [KernelSkill: A Multi-Agent Framework for GPU Kernel Optimization](https://mlsanity.com/paper/2603.10085/) (2026)
- [VibeTensor: System Software for Deep Learning, Fully Generated by AI Agents](https://mlsanity.com/paper/2601.16238/) (2026)
- [ML-Triton, A Multi-Level Compilation and Language Extension to Triton GPU Programming](https://mlsanity.com/paper/2503.14985/) (2025)

## Built on by (most cited first, 2 of 2)

- [From PyTorch to the NPU: LLM-Agent-Driven Model Conversion Across Heterogeneous Inference Runtimes](https://mlsanity.com/paper/2609.27249/) (2026)
- [D2K-Bench: Can LLM Agents Turn Expert Designs into Efficient GPU Kernels?](https://mlsanity.com/paper/2610.03226/) (2026)

## Abstract (arXiv)

Porting deep learning algorithms to new hardware accelerators requires developers to repeatedly apply the same low-level optimizations -- quantization, memory access coalescing, tile size tuning, and architecture-specific workarounds -- to every Triton kernel in their code-base. This manual, repetitive effort is a major bottleneck: each kernel demands the same cycle of trial-and-error profiling against hardware constraints that vary across devices, yet the underlying optimization patterns remain largely consistent. We present Xe-Forge, a multi-stage LLM-powered pipeline that automates this process for Intel GPU. Given a functionally correct Triton kernel, the system applies up to nine optimization stages -- from algorithmic restructuring and operator fusion through block pointer modernization, GPU-specific tuning, and open-ended discovery -- each driven by a Chain-of-Verification-and-Refinement (CoVeR) agent that generates candidates, validates them on real hardware, and iterates on failures. A curated knowledge base encodes Intel GPU constraints (power-of-two warp counts, GRF modes, SLM sizing) that are absent from LLM training data, keeping the model within architecturally valid bounds. We evaluate Xe-Forge on 97 Level-2 KernelBench kernels and Flash Attention on the Intel Arc Pro B70, achieving a 1.17x geometric mean speedup over PyTorch eager with 67% of kernels improving, nine kernels exceeding 5x (up to 82x), and 2--13.3x speedups on Flash Attention across all tested configurations without regression -- demonstrating that structured domain knowledge with hardware-in-the-loop verification can systematically eliminate the repetitive porting effort that currently gates algorithm deployment on new accelerators.