-
rayon
work-stealing parallelism for Rust
-
profiling
very thin abstraction over other profiler crates
-
divan
Statistically-comfy benchmarking library
-
mimalloc
Performance and security oriented drop-in allocator
-
blstrs
High performance implementation of BLS12 381
-
gix-discover
Discover git repositories and check if a directory is a git repository
-
tokio-websockets
High performance, strict, tokio-util based WebSockets implementation
-
nucleo
plug and play high performance fuzzy matcher
-
simdeez
SIMD library to abstract over different instruction sets and widths
-
ada-url
Fast WHATWG Compliant URL parser
-
oha
Ohayou(おはよう), HTTP load generator, inspired by rakyll/hey with tui animation
-
libmimalloc-sys
Sys crate wrapping the mimalloc allocator
-
trueno
High-performance SIMD compute library with GPU support for matrix operations
-
kode-bridge
Modern HTTP Over IPC library for Rust with both client and server support (Unix sockets, Windows named pipes)
-
gungraun
High-precision, one-shot and consistent benchmarking framework/harness for Rust. All Valgrind tools at your fingertips.
-
inline-array
efficient immutable inlinable byte array
-
xilem
A next-generation cross-platform Rust UI framework
-
mimalloc-safe
Performance and security oriented drop-in allocator
-
prmt
Ultra-fast, customizable shell prompt generator with zero-copy parsing
-
precord
Command line tool for recording process or system performance data
-
reactive-mutiny
async Event-Driven Reactive Library with advanced & optimized containers (channels) and Stream executors
-
concurrent-map
a lock-free linearizable B+ tree
-
patina
Common types and functionality used in UEFI development
-
fast-down-cli
超级快的下载器命令行界面
-
toprs
Cross platform memory focused top
-
codspeed-divan-compat-walltime
A temporary compatibility layer for CodSpeed to use Divan's walltime entrypoint
-
native_model
A thin wrapper around serialized data which add information of identity and version
-
cargo-multivers
Cargo subcommand to build multiple versions of the same binary, each with a different CPU features set, merged into a single portable optimized binary
-
lite-sync
Fast, lightweight async primitives: SPSC channel, oneshot, notify, and atomic waker
-
cargo-perf
Preventive performance analysis for Rust - catch anti-patterns before production
-
logly
High-performance, structured logging library with async support, rotation, filtering, and GPU/CPU optimization
-
gdt-cpus
Game Developer's Toolkit for CPU Management
-
momento-functions-host
Host interface support crate for Momento Functions
-
locus-cli
A configurable CPU stress testing tool with multiple computational workloads
-
perf-event-open
Full-featured high-level wrapper for the perf_event_open system call
-
iyes_perf_ui
Customizable Performance/Debug Overlay for Bevy UI
-
profile-inspect
Analyze V8 CPU and heap profiles from Node.js/Chrome DevTools
-
arena-b
Production-grade bump allocator with lock-free, slab, and virtual-memory tooling for parsers, game engines, and request-scoped services
-
pzsh
Performance-first shell framework with sub-10ms startup
-
rusted-ring
LMAX inspired Lock-free ring buffers with cache-aligned memory pools for high-performance event systems
-
rcp-tools-rrm
Fast file removal tool - efficiently removes large filesets (similar to rm -rf but much faster)
-
blstrs_plus
High performance implementation of BLS12 381
-
mobench-sdk
Mobile benchmarking SDK for Rust - run benchmarks on real devices
-
smartbuf
A high-performance buffered reader with background thread pre-fetching and full seek support
-
rcp-tools-rlink
Fast hard-linking tool - efficiently creates hard links for large filesets (inspired by pcp/dsync)
-
protosocket
Message-oriented nonblocking tcp stream
-
rcp-tools-rcmp
Fast file comparison tool - efficiently compares metadata across large filesets
-
feoxdb
Iron-oxide fast embedded database - nanosecond-level key-value storage
-
atomptr
A safe copy-on-write wrapper around
AtomicPtrwith some extra tricks -
ebr
epoch-based reclamation with low cacheline ping-pong
-
cargo-simplebench
minimalist microbenchmarking framework with clear regression detection
-
netperf
A network performance measurement tool
-
orasort
High-performance, cache-efficient sorting library optimized for common prefixes
-
fastset
Fast set implementation for dense, bounded integer collections, optimized for quick updates and access
-
oxcache
A high-performance multi-level cache library for Rust with L1 (memory) and L2 (Redis) caching
-
evno
A high-performance event bus for asynchronous tasks and event-driven systems
-
rat_engine
高性能Rust HTTP服务器引擎，提供Flask风格的API和自动硬件优化功能
-
base64-turbo
The fastest memory-safe Base64 engine. Formally verified (Kani/MIRI). Features runtime AVX2/AVX512 detection, zero-allocation API, and robust no_std support.
-
zeropool
High-performance buffer pool with constant-time allocation, thread-safe operations, and 5x speedup over bytes crate
-
zebo
Write data on FS in a performance way
-
keen-retry
-- yet powerful -- zero-cost-abstractions & zero-copy lib for error handling & recovery
-
simdly
🚀 High-performance Rust library leveraging SIMD and Rayon for fast computations
-
fastalloc
High-performance memory pooling library with type-safe handles, predictable latency, and zero fragmentation. Perfect for game engines, real-time systems, and high-churn workloads.
-
ugnos
A high-performance, concurrent time-series database core written in Rust, designed for efficient IoT data ingestion, real-time analytics, and monitoring
-
efficient-sm2
China's Standards of Signature Algorithms (SM2)
-
inqjet
Ultra-fast, low-latency logging for Rust applications
-
mobench
Mobile benchmarking CLI for Rust - Run benchmarks on real Android and iOS devices
-
level-runtime
Conveniences for tokio thread-per-core runtime configuration
-
momento-functions
Support crate for Momento Functions
-
signvec
Vector implementation for fast, sign-based manipulation of dynamic collections
-
ttlog
A comprehensive high-performance tracing and logging system
-
mimalloc-rspack
Performance and security oriented drop-in allocator
-
clickhouse-datafusion
High-performance ClickHouse integration for Apache DataFusion with federation support
-
router_prefilter
Fast prefix-based prefiltering for router pattern matching
-
cargo-samply
A cargo subcommand to automate the process of running samply for project binaries
-
rcp-tools-rcp
Fast file operations tools - rcp (copy) and rcpd (remote copy daemon)
-
rperf3-rs
High-performance network throughput measurement tool, inspired by iperf3
-
nam-blstrs
High performance implementation of BLS12 381
-
momento-functions-log
Log adapter for momento Functions
-
wasps-with-bazookas
Distributed load testing tool - like bees with machine guns, but way more power!
-
ruvector-memopt
Intelligent cross-platform memory optimizer with neural learning capabilities for smart optimization decisions
-
bufjson
No frills, low-alloc, low-copy JSON lexer/parser for fast stream-oriented parsing
-
timberio/file-source
A high-performance observability data pipeline
-
system-analysis
A comprehensive Rust library for analyzing system capabilities, workload requirements, and optimal resource allocation
-
predate
Catch bugs and performance regressions through automated system testing
-
tilesort
A sorting algorithm optimized for datasets with pre-sorted contiguous blocks (tiles)
-
clustering
easy way to perform kmeans clustering on arbitrary data
-
cntryl-stress
Lightweight single-shot benchmark runner for system-level stress tests
-
bevy_perf_ui
Customizable Performance/Debug Overlay for Bevy UI
-
tcmalloc-better
wrapper over Google's TCMalloc memory allocator
-
network-protocol
Secure, high-performance protocol core with backpressure control, structured logging, timeout handling, TLS support, and comprehensive benchmarking for robust Rust networked applications and services
-
baelNET
BaelNET: Post-Quantum Hybrid Encryption with high throughput and authenticated wire format
-
siteprobe
CLI tool to fetch URLs from sitemap.xml, check their existence, and generate performance reports
-
premix-core
A Zero-Overhead, Type-Safe ORM for Rust that runs optimized SQL
-
gnu-sort
High-performance Rust implementation of GNU sort with zero-copy operations, SIMD optimization, and parallel processing
-
tango-bench
Tango benchmarking harness
-
axiom_mem
A stable and high-performance low-level memory management library for Rust, providing fixed-capacity pools and deterministic allocation for latency-sensitive systems
-
relay-bp
relay belief propagation for qLDPC codes
-
maker_web
Security-first, high-performance, zero-allocation HTTP server for microservices
-
tempfile-fast
Support for Linux-specific tempfile extensions
-
swb-sys-monitor
为嵌入式设备高并发监控场景设计的极简网页资源占用显示系统
-
rat_logger
具有生产者消费者架构、异步文件操作和分层配置系统的高性能日志库
-
parcode
A high-performance, lazy load and parallelized caching library for complex Rust data structures
-
b10
Lossless handling of decimal numbers at maximum performance
-
moos
Memory-Optimized Objects and Strings (MOOS)
-
inference-lab
High-performance LLM inference simulator for analyzing serving systems
-
sql_query_analyzer
Static analysis tool for SQL queries with 20 built-in rules for performance, security, and style
-
startled
CLI tool for benchmarking Lambda functions
-
embeddenator-testkit
Comprehensive testing utilities and performance benchmarking for embeddenator VSA operations
-
getattrlistbulk
Safe Rust bindings for macOS getattrlistbulk() system call for high-performance directory enumeration
-
double_linked_list
High-performance doubly-linked list with smart pointer support and memory pool optimization
-
gtfs-guru
Command-line interface for GTFS Guru validator
-
graus-db
A high-performance, thread-safe key-value embedded data store
-
thyme-sql
Performance benchmark tool for Postgres queries
-
bench_diff
Reliably compare the latencies of two functions/closures
-
fast-yaml-parallel
Multi-threaded YAML processing with work-stealing parallelism
-
datavalue-rs
High-performance Arena-based alternative to serde_json::Value
-
nanobit
Ultra-fast binary serialization with multi-format compression and zero-copy deserialization
-
momento-protos
Generated GRPC and Protocol Buffers types for Momento clients
-
zero-pool
High-performance thread pool with consistent low-latency task dispatch
-
patina_performance
Performance measurement infrastructure
-
static-id
A memory-efficient library for handling interned identifiers
-
hft-benchmarks
High-precision benchmarking tools for high-frequency trading systems with nanosecond-level timing accuracy
-
rjprof-cli
Command-line interface for rjprof Java profiler
-
mix
ui library cooked up in half a day! (expect extra salt)
-
zaw
Zero-allocation WebAssembly communication protocol for Rust
-
domain-key
High-performance, domain-driven, type-safe key system for Rust
-
rmbrr
Fast parallel directory deletion with cross-platform support
-
ordsearch
A data structure for efficient lower-bound lookups
-
quantum_log
High-performance asynchronous logging framework based on tracing ecosystem
-
harness
Precise and reproducible benchmarking
-
spaa_parse
Parser and writer for SPAA (Stack Profile for Agentic Analysis) files
-
k-lock
A fast mutex for short critical sections
-
dash-em
Enterprise-Grade Em-Dash Removal Library — SIMD-Accelerated String Processing
-
divan-macros
Macros for Divan, a statistically-comfy benchmarking library
-
fast-steal
特别快的多线程任务调度器，支持超细颗粒度的任务窃取
-
batched
rust macro util for batching expensive operations
-
integrate
Small, lightweight Rust library for performing numerical integration
-
json2bin
A fast jsonl to RWKV binidx converter in Rust
-
spillway
High throughput, safe, async MPSC channel
-
static_interner
Similar to
internmentcrate, but with interface and performance tweaks -
fast-pull
Pull everything fast
-
qr-tester
A CLI tool to scan QR codes from images and benchmark performance
-
benchkit
Lightweight benchmarking toolkit focused on practical performance analysis and report generation. Non-restrictive alternative to criterion, designed for easy integration and markdown report generation
-
gpui_rn_perf
Performance measurement utilities (fork for React Native GPUI)
-
pistonite-cu
Battery-included common utils to speed up development of rust tools
-
jist
Welcome to Jist a lightweight and efficient programming language built in Rust, designed for simplicity, flexibility, and fast performance with a clean and readable syntax. Github: https://github.com/jon429r/JistR
-
melange_db
基于 sled 架构深度优化的下一代高性能嵌入式数据库，支持 ARM64 NEON SIMD 优化、多级缓存和布隆过滤器
-
peak-mem
Lightweight memory usage monitor for any process
-
typestorm
A cool terminal typing test CLI written in rust
-
error_king
CLI for 3RROR_K1NG - Website Roast Machine. Get your website brutally roasted with AI-powered security, performance, SEO & accessibility audits.
-
lightning-log
Ultra-fast zero-allocation logging for high-frequency trading and low-latency systems
-
fulgurance
A blazing-fast, adaptive prefetching and caching library for Rust
-
tagged_dispatch
Memory efficient trait dispatch using tagged pointers
-
gpui_perf
measuring Zed test performance, with too many Clippy lints
-
crisp-cli
Crisp CLI — website audit tool. Security, performance, SEO & accessibility audits from your terminal.
-
slipstream
SIMD library usable by the masses
-
rat_memcache
高性能 Memcached 协议兼容服务器，支持双层缓存和持久化存储
-
stack-arena
A fast, stack-like arena allocator for efficient memory management, implemented in Rust
-
bytesbuf_io
Asynchronous I/O abstractions expressed via
bytesbuftypes -
numaperf-topo
NUMA topology discovery and CPU locality mapping
-
json-schema-validator-core
Lightning-fast JSON schema validation library with custom error messages and multi-language bindings
-
webbuf
fast base64 and hex encoding/decoding
-
hive-router-internal
GraphQL Hive Router internal crate
-
protosocket-prost
Message-oriented nonblocking tcp stream - protocol buffers bindings
-
uninit-read
A marker trait and utilities for safe, high-performance reads into uninitialized buffers
-
better_mimalloc_rs
A mimalloc wrapper that exposes tuning knobs and tracks the dev branch
-
temporal-neural-solver
Ultra-fast neural network inference with sub-microsecond latency
-
hash_on_write
A wrapper for storing hash results to avoid running costly hash functions multiple times without modifying the value
-
fastcache
A performant but not-so-accurate time and capacity based cache for Rust
-
pore
a performance oriented reimplementation of repo
-
rat_congestion
High-performance congestion control algorithms for network protocols
-
granular-metrics
Zero-middleware metrics counter that emits per-key and aggregate RPS/RPM snapshots every second
-
numaperf-affinity
Thread pinning and CPU affinity management
-
fast-math
Fast, approximate versions of mathematical functions
-
http-timings
measure the key HTTP timings from the development tools
-
protosocket-server
Message-oriented nonblocking tcp stream - server tools
-
momento-functions-wit
Internal support crate for Momento Functions
-
fast-layer
WIP: A fast WebAssembly-based layer for high-performance MapLibre/Mapbox
-
bael_crypt
The Bael Crypt Protocol
-
numaperf-mem
NUMA-aware memory allocation and placement policies
-
heuristics
Rules of thumb to improve Rust code
-
bma-benchmark
Benchmark for Rust and humans
-
fast-decision
High-performance rule engine
-
zon-format
6.2x faster than JSON. A zero-copy, memory-mapped data format for high-performance systems.
-
rmemstore
client for rmemstored
-
kawa-storage
High-performance storage engine for Kawa message broker
-
numaperf-core
Core types and error handling for the numaperf NUMA runtime
-
coreml-rs
CoreML bindings for Rust using swift-bridge to maximize performance
-
better_mimalloc_sys
Sys crate wrapping mimalloc (dev-branch fork for better_mimalloc_rs)
-
numaperf-sharded
Per-node sharded data structures for NUMA systems
-
prs-rs
High performance compressor/decompressor for the SEGA PRS Compression scheme
-
rapid-delete-lib
A high-performance file deletion library
-
giga-chess
chess library built for performance, handling game logic and legal/best move generation
-
numaperf-io
Device locality discovery for NICs and storage
-
macrotime
efficient timing
-
tiverse-mmap
Modern, safe, and ergonomic memory-mapped file I/O library with zero unsafe in public API
-
const_init_build
Get Rust constant variables from configuration file at build time
-
numaperf
NUMA-first runtime for latency-critical Rust applications
-
conda_curation
Reduce conda repodata to enforce policy and speed up solves. Alpha software.
-
lake
high-octane memory pool with direct access, checkpoints, rollback, and zero-allocation droplets. No GC. No overhead. Just you and the bytes — total control. Allocate like a cyber-samurai.
-
axonml-profile
Profiling tools for the Axonml ML framework
-
libdd-log
A bridge from Rust to {language} logging
-
axonml-fusion
Kernel fusion optimization for the Axonml ML framework
-
ninjatracing
Rust port of ninjatracing: converts Ninja build logs to Chrome Tracing format
-
hedged-rpc-client
High-performance Solana RPC client with request hedging for tail latency elimination
-
simple-cacher
A high-performance, flexible caching library with custom matching capabilities and automatic expiration
-
gpqa-example
Example implementation of GPQA (Graduate-Level Google-Proof Q&A) evaluation tasks
-
learned-partition-sort
A high-performance distribution-based sorting algorithm that learns data patterns to achieve O(N) complexity
-
malloc-best-effort
wrapper over Google's TCMalloc and Microsoft's MIMalloc memory allocators
-
ppd
Interact with power-profiles-daemon with Rust
-
baseten_performance_client_core
High performance HTTP client for Baseten.co and other APIs
-
oshash
https://pypi.org/project/oshash/ in Rust
-
bit-parallel-search
Blazing fast string search using bit-parallel algorithms - up to 8x faster than naive search
-
hashmap-mem
Fast, low-overhead in-memory hashmap implementation optimized for performance using fxhash
-
mu_uefi_perf_timer
Performance timer support
-
coreml-rs-fork
CoreML bindings for Rust using swift-bridge to maximize performance
-
perf_monitor
A toolkit designed to be a foundation for applications to monitor their performance
-
protosocket-messagepack
Message-oriented nonblocking tcp stream - messagepack serde bindings
-
rayon-scan
A parallel prefix scan function for ParallelIterator
-
dropslot
A high-performance publish-subscribe library with latest-only delivery semantics
-
parallel_bzip2_decoder
High-performance parallel bzip2 decompression library
-
flashcron
A lightning-fast, ultra-efficient cron daemon written in Rust - Schedule tasks at the speed of light
-
fact-wasm-core
FACT WASM Core - High-performance WebAssembly cognitive processing engine with advanced caching and template optimization
-
quick-bool
A lock-free boolean implementation using atomic operations
-
cache-size
finding your L1/L2/L3 cache sizes
-
radix256_sort
A high-performance Radix Sort implementation using base-256 for u32
-
k-cache
internal cache implementation for rmemstore
-
math-example
mathematical evaluation example demonstrating the nanoeval framework
-
runmod
libary to speed up development
-
win-rmdir-fast
Hyper-optimized Windows directory deletion tool
-
auto-allocator
Zero-configuration automatic memory allocator selection based on runtime hardware analysis and performance research
-
freedesktop-icon
Rust implementations of the freedesktop standards
-
nypc-perf
Performance calculation of NYPC
-
perfcnt
configure and read hardware performance counters in rust
-
des-cartes
DesCartes: A deterministic, replayable, discrete-event simulator for Rust
-
ambient_profiling
very thin abstraction over other profiler crates
-
haagenti-simd
SIMD-accelerated primitives for Haagenti
-
quote_into
Easily & efficiently generate code by quoting it in a macro
-
system-profile
Cached system profile information for runtime optimization decisions
-
bevy_full_throttle
Enable CPU performance mode for Bevy games
-
randstream
Reproducible Random Stream Generator and Validator
-
yep-cache-line-size
A small library for retrieving cache line sizes. Forked from lovesegfault/cache-size.
-
llm-latency-lens-core
Core types and timing engine for LLM Latency Lens
Try searching with DuckDuckGo.