Open Source Performance Tools
Discover 54 open source Performance repositories from GitHub, each with an analysis of what it does, key features, use cases and alternatives. Performance projects here are most often combined with Library, Python and Developer Tools. Last updated October 4, 2026.
54 repositories · updated October 4, 2026
es-toolkit: Modern JavaScript Utility Functions
es-toolkit is a TypeScript-first JavaScript utility library built for smaller bundles and performance in modern environments. Use it for everyday helpers or as a compatibility-oriented alternative when replacing Lodash.
insanely-fast-whisper: Transcribe Audio with Whisper
A command-line tool for running Whisper speech recognition on your own NVIDIA GPU or Apple Silicon Mac. It combines batching and optional Flash Attention to speed up transcription, with options for timestamps, translation, and speaker diarization.
flash-attention: Compute Exact Attention Faster and with Less Memory
FlashAttention provides GPU-optimized implementations of exact attention for deep learning, reducing memory use while improving speed. It is aimed at practitioners training or serving Transformer models who can build and run GPU kernels.
ZLUDA: Run CUDA Applications on Non-NVIDIA GPUs
ZLUDA aims to let unmodified CUDA applications run on non-NVIDIA GPUs as a drop-in CUDA replacement. It is for developers and users who want to try CUDA software on supported non-NVIDIA hardware; check the project documentation for setup and compatibility details.
memray: Profile Python Memory Allocations
Memray traces memory allocations in Python programs, including native C and C++ code, to help explain high memory use and find leaks. Use it to inspect allocation call stacks through command-line reports or its Python API.
faiss: Search and Cluster Dense Vectors Efficiently
Faiss provides indexing methods for similarity search and clustering over dense vectors, with trade-offs between search speed, accuracy, and memory use. It suits teams building vector retrieval or large-scale nearest-neighbor systems in C++ or Python.
zed: A High-Performance Multiplayer Code Editor
Zed is a Rust-based code editor designed for fast, responsive editing and multiplayer collaboration. It suits developers who want a desktop editor for individual work or shared coding sessions across macOS, Linux, and Windows.
pnpm: Manage JavaScript Dependencies Efficiently
pnpm is a Node.js package manager that links packages from shared content-addressable storage to reduce duplicated files and speed up installs. It suits JavaScript projects, especially monorepos, that need deterministic dependency management.
KeyDB: Run a Multithreaded Redis-Compatible Database
KeyDB is a high-performance, multithreaded fork of Redis for workloads that need more throughput or features such as active replication. Its Redis protocol compatibility makes it a potential drop-in replacement, with Linux-focused build and deployment requirements.
optimum: Optimize Model Training and Inference on Target Hardware
Hugging Face Optimum adds tools for optimizing model training and inference across hardware backends. It suits teams using Transformers, Diffusers, TIMM, or Sentence Transformers who need hardware-specific deployment or training workflows.
s-tui: Monitor and Stress-Test CPUs in the Terminal
s-tui is a terminal interface for monitoring CPU temperature, frequency, power, and utilization, with built-in stress testing. It suits Linux users who want to observe CPU behavior or check for thermal throttling without a graphical desktop.
cytoolz: Speed Up Python Functional Utilities
cytoolz provides Cython implementations of toolz utilities for working with iterables, functions, and dictionaries. Use it when toolz-compatible APIs and iterator-friendly processing need more speed, or when Cython projects need its C API.