Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Review automated kernel generation in the era of LLMs
| Date | Stars |
|---|---|
| 2026-07-31 | 278 |
| 2026-08-11 | 285 |
| 2026-08-18 | 292 |
| 2026-08-19 | 291 |
| 2026-08-20 | 292 |
| 2026-08-22 | 293 |
| 2026-08-23 | 296 |
| 2026-08-24 | 300 |
| 2026-08-27 | 302 |
| 2026-08-28 | 304 |
| 2026-08-31 | 305 |
| 2026-09-01 | 307 |
| 2026-09-08 | 308 |
| 2026-09-09 | 308 |
| 2026-09-10 | 309 |
| 2026-09-16 | 310 |
| 2026-09-17 | 310 |
| 2026-09-20 | 310 |
Today
— stars today
This week
+1 stars this week
This month
+18 stars this month
Momentum
0.0
growth rate 0.32%/day
<center> <h1 align="center"> :cherries: Awesome LLM-Driven Kernel Generation </h1> <h2 align="center">🔥 <a href="https://arxiv.org/abs/2601.15727">Paper</a> 🔥</h2>  </center> The integration of Large Language Models (LLMs) and agentic systems marks a pivotal shift in high-performance computing, transforming kernel engineering from a labor-intensive, expert-dependent process into a scalable, automated workflow. To provide a systematic perspective on this rapidly evolving field, we summarize the related literature below. Note that the works are organized according to the taxonomy proposed in the survey. We categorized these researches into four main streams: # 📌 Table of Content (ToC) - [LLM4Kernel](#LLM4Kernel) - [Agent4Kernel](#Agent4Kernel) - [Datasets](#Datasets) - [Benchmarks](#Benchmarks) # LLM4Kernel <center>  </center> Applying LLMs to kernel synthesis presents universal challenges in correctness and performance-sensitive structuring across diverse programming abstractions. Addressing these complexities, this section reviews the two principal post-training methodologies that dominate current research: supervised fine-tuning and reinforcement learning. ## SFT \[06/2025] KernelLLM: Making Kernel Development More Accessible [\[link\]](https://huggingface.co/facebook/KernelLLM) \[10/2025] ConCuR: Conciseness Makes State-of-the-Art Kernel Generation [\[paper\]](https://arxiv.org/abs/2510.07356) \[03/2026] InCoder-32B: Code Foundation Model for Industrial Scenarios [\[paper\]](https://arxiv.org/pdf/2603.16790) | [\[code\]](https://github.com/CSJianYang/Industrial-Coder) \[04/2026] InCoder-32B-Thinking: Industrial Code World Model for Thinking [\[paper\]](https://arxiv.org/pdf/2604.03144) | [\[code\]](https://github.com/CSJianYang/Industrial-Coder) ## RL \[07/2025] AutoTriton: Automatic Triton Programming with Reinforcement Learning in LLMs [\[paper\]](https://arxiv.org/abs/2507.05687) \[07/2025] CUDA-L1: Improving CUDA Optimization via Contrastive Reinforcement Learning [\[paper\]](https://arxiv.org/abs/2507.14111) | [\[code\]](https://github.com/deepreinforce-ai/CUDA-L1) \[07/2025] Kevin: Multi-Turn RL for Generating CUDA Kernels [\[paper\]](https://arxiv.org/abs/2507.11948) \[09/2025] Mastering Sparse CUDA Generation through Pretrained Models and Deep Reinforcement Learning [\[paper\]](https://openreview.net/forum?id=VdLEaGPYWT) \[10/2025] TritonRL: Training LLMs to Think and Code Triton Without Cheating [\[paper\]](https://arxiv.org/abs/2510.17891) \[11/2025] QiMeng-Kernel: Macro-Thinking Micro-Coding Paradigm for LLM-Based High-Performance GPU Kernel Generation [\[paper\]](https://arxiv.org/abs/2511.20100) \[12/2025] CUDA-L2: Surpassing cuBLAS Performance for Matrix Multiplication through Reinforcement Learning [\[paper\]](https://arxiv.org/abs/2512.02551) \[01/2026] AscendKernelGen: A Systematic Study of LLM-Based Kernel Generation for Neural Processing Units [\[paper\]](https://arxiv.org/abs/2601.07160) \[02/2026] Dr. Kernel: Reinforcement Learning Done Right for Triton Kernel Generations [\[paper\]](https://arxiv.org/pdf/2602.05885) | [\[code\]](https://github.com/hkust-nlp/KernelGYM) \[02/2026] Improving HPC Code Generation Capability of LLMs via Online Reinforcement Learning with Real-Machine Benchmark Rewards [\[paper\]](https://arxiv.org/pdf/2602.12049) \[02/2026] CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation [\[paper\]](https://arxiv.org/pdf/2602.24286) | [\[code\]](https://github.com/BytedTsinghua-SIA/CUDA-Agent) \[03/2026] Kernel-Smith: A Unified Recipe for Evolutionary Kernel Optimization [\[paper\]](https://arxiv.org/abs/2603.28342) # Agent4Kernel <center>  </center> While foundational LLMs are often limited to static, one-pass inference, agentic systems introduce an autonomous, closed-loop paradigm characterized by iterative planni
Excerpt of 24,240 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:40ad5dd6591b04b1, llm:Repository title and description: 'awesome-LLM-driven-kernel-generation' — 'Review automated kernel generation in the era of LLMs'. Appears to be an 'awesome' list reviewing LLM-driven kernel generation techniques/tools.
matched fp:40ad5dd6591b04b1, llm:Repository title and description: 'awesome-LLM-driven-kernel-generation' — 'Review automated kernel generation in the era of LLMs'. Appears to be an 'awesome' list reviewing LLM-driven kernel generation techniques/tools.
matched fp:40ad5dd6591b04b1, llm:Repository title and description: 'awesome-LLM-driven-kernel-generation' — 'Review automated kernel generation in the era of LLMs'. Appears to be an 'awesome' list reviewing LLM-driven kernel generation techniques/tools.