llvm-propeller
Recompilation optimizer
An optimization tool for improving the performance of large-scale software recompilation using profiling and relinking techniques.
PROPELLER: Profile Guided Optimizing Large Scale LLVM-based Relinker
381 stars
16 watching
36 forks
Language: C++
last commit: almost 2 years agoLinked from 1 awesome list
caerind/awesomecppgamedevRelated projects:
| Repository | Description | Stars |
|---|---|---|
 torch/optim | A package that provides tools for finding and improving the parameters of a mathematical model to optimize its performance | 196 |
 lge-arc-advancedai/auptimizer | Automates model building and deployment process by optimizing hyperparameters and compressing models for edge computing. | 200 |
 alibaba/conv-llava | This project presents an optimization technique for large-scale image models to reduce computational requirements while maintaining performance. | 106 |
| google/jaxopt | An open-source project providing hardware accelerated, batchable and differentiable optimizers in JAX for deep learning. | 941 |
 lyogavin/anima | An optimization technique for large language models allowing them to run on limited hardware resources without significant performance loss. | 9 |
 google-research/sputnik | A library of optimized GPU kernels for sparse matrix operations used in deep learning. | 248 |
 kvcache-ai/ktransformers | A flexible framework for LLM inference optimizations with support for multiple models and architectures | 771 |
 slorber/rereduce | A Redux reducer library that optimizes state management by memoizing and linking dependencies between reducers. | 163 |
 jiangoforit/yellowfin_pytorch | An optimizer that automatically tunes momentum and learning rate using local quadratic approximation. | 287 |
 pnnl/soda-opt | A tool that extracts and optimizes high-level code snippets into LLVM IR using MLIR for high-level synthesis | 37 |
 claesenm/optunity | A collection of algorithms for hyperparameter optimization in machine learning models | 417 |
 alan-fgr/cullminator9000 | A high-performance software optimization tool for removing unnecessary geometric data in 3D rendering pipelines | 35 |
 ibm/reprogrammble-fl | Improves utility-privacy tradeoff in federated learning by reprogramming models to balance data utility and user privacy. | 5 |
 tensorzero/tensorzero | A tool for optimizing large language models by collecting feedback and metrics to improve their performance over time | 1,245 |
 fisheyehq/module-url-rewrite-optimiser | A Magento module that prevents unnecessary URL rewrites for products with category paths when the 'Use Categories Path for Product URLs' setting is disabled. | 59 |