#genAI: new offers in the open-source arena ▪️Meta has released Meta LLM Compiler. #LLMs are already widely used in software engineering for tasks such as ▪️generating code and documentation, ▪️translating code between programming languages, ▪️supporting testing and writing unit tests, and ▪️detecting and fixing bugs. In March 2024, researchers from Meta AI and Rice University introduced ▪️compiler optimization powered by LLMs with compiler feedback to optimize the code size of LLVM assembly. 📁 Article „Compiler generated feedback for Large Language Models“ on arXiv: https://lnkd.in/dJ3vzaBQ These developments facilitate that large open-source code datasets and Github-enabled models such as CodeLlama, ChatGPT, and Codex can help to develop a deeper understanding of various programming languages.
Today we’re releasing Meta LLM Compiler, a family of models built on Meta Code Llama with additional code optimization and compiler capabilities. The models achieve state-of-the-art results on optimization of code size and disassembly tasks. Hugging Face repo ➡️ https://go.fb.me/tdd3dw Research paper ➡️ https://go.fb.me/hvfnam LLM Compiler can emulate the compiler, predict optimal passes for code size, and disassemble code. It can be fine-tuned for new optimizations and compiler tasks. This work shows that AI is learning to optimize code and can assist compiler experts in identifying opportunities to apply optimizations. We believe this work could have an impact ranging from use in optimization for individual developer environments to inclusion in a compiler such as LLVM. We’re releasing LLM Compiler 7B & 13B models under a permissive license for both research and commercial use in the hopes of making it easier for developers and researchers alike to leverage this in their work and carry forward new research in this highly impactful space.
Project „AI Opportunities“ Research Analyst at Deutsche Bank (Technology and AI)
2mohttps://www.marktechpost.com/2024/03/26/the-idea-of-compiler-generated-feedback-for-large-language-models/