# Kernel Forge: An Agent Harness for LLM-based Generation and Optimization of CUDA Kernels

arXiv: https://arxiv.org/abs/2607.24762 · HTML: https://mlsanity.com/paper/2607.24762/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 22 of its 42 references that are in the map.
- First posted on arXiv 2 Jun 2026.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 22 of 22)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [In-Datacenter Performance Analysis of a Tensor Processing Unit](https://mlsanity.com/paper/1704.04760/) (2017)
- [Efficient Processing of Deep Neural Networks: A Tutorial and Survey](https://mlsanity.com/paper/1703.09039/) (2017)
- [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)
- [Learning to Optimize Tensor Programs](https://mlsanity.com/paper/1805.08166/) (2018)
- [TritonBench: Benchmarking Large Language Model Capabilities for Generating Triton Operators](https://mlsanity.com/paper/2502.14752/) (2025)
- [Towards Robust Agentic CUDA Kernel Benchmarking, Verification, and Optimization](https://mlsanity.com/paper/2509.14279/) (2025)
- [Astra: A Multi-Agent System for GPU Kernel Performance Optimization](https://mlsanity.com/paper/2509.07506/) (2025)
- [CudaForge: An Agent Framework with Hardware Feedback for CUDA Kernel Optimization](https://mlsanity.com/paper/2511.01884/) (2025)
- [MultiKernelBench: A Multi-Platform Benchmark for Kernel Generation](https://mlsanity.com/paper/2507.17773/) (2025)
- [Geak: Introducing Triton Kernel AI Agent & Evaluation Benchmarks](https://mlsanity.com/paper/2507.23194/) (2025)
- [FlashInfer-Bench: Building the Virtuous Cycle for AI-driven LLM Systems](https://mlsanity.com/paper/2601.00227/) (2026)
- [Autocomp: A Powerful and Portable Code Optimizer for Tensor Accelerators](https://mlsanity.com/paper/2505.18574/) (2025)
- [KernelBenchX: A Comprehensive Benchmark for Evaluating LLM-Generated GPU Kernels](https://mlsanity.com/paper/2605.04956/) (2026)
- [CUDABench: Benchmarking LLMs for Text-to-CUDA Generation](https://mlsanity.com/paper/2603.02236/) (2026)
- [A Systematic Characterization of LLM Inference on GPUs](https://mlsanity.com/paper/2512.01644/) (2025)

## Built on by (most cited first, 1 of 1)

- [iCoder-27B: Recursive AI-Led Development of Frontier Industrial Coding Model](https://mlsanity.com/paper/2609.29626/) (2026)

## Abstract (arXiv)

Machine learning models are increasingly embedded in everyday software, and most of their runtime is spent in a small set of compute kernels such as matrix multiplication, convolution, and normalization. Optimizing these kernels is one of the most direct ways to reduce latency and cost, but it has traditionally required expert engineers to hand-write low-level GPU code. Agentic systems built on large language models (LLMs) can now generate and optimize kernels with far less human effort, yet existing tools are largely evaluated on randomly generated tensors and isolated kernels, emit standalone CUDA code that developers must manually reintegrate, mostly target only LLM PyTorch models, and offer limited support for inspecting and debugging results. We present Kernel Forge, an open-source, end-to-end agentic harness that accepts any unmodified PyTorch model in place. Kernel Forge supports vision, diffusion, and LLM workloads, uses Monte Carlo Tree Search (MCTS) to explore multiple optimization paths rather than a single linear refinement chain, and ships with a graphical user interface for monitoring progress, inspecting candidate kernels, and debugging failures. We evaluate Kernel Forge on four PyTorch models spanning vision, diffusion, and LLM workloads on an NVIDIA DGX Spark with GB10 GPU. With only 50 optimization iterations per kernel, it optimizes 14 kernels to outperform PyTorch eager mode, reaching 1.52× on adaptiveavgpool2d in ResNet-50, 1.70× on groupnorm in Stable Diffusion 3.5 Medium, 2.83× on softmax in Gemma 4 E2B, and 1.54× on softmax in Qwen 3.5 35B-A3B.