Low Latency Python Packages

python-socketio

Python Socket.IO server and client

databento-dbn

Databento Binary Encoding (DBN) - Fast message encoding and storage format for market data

dxcam

A Python high-performance screen capture library for Windows using Desktop Duplication API - Updated 2026

thetadatadx

No-JVM ThetaData Terminal — native Rust SDK for direct MDDS (gRPC) and FPSS (TCP) market data access

znsocket

Python implementation of a Redis-compatible API using websockets.

dbz-python

Databento Binary Encoding (DBN) - Fast message encoding and storage format for market data

fastq-redis

Ultra-fast job queue written in C. Drop-in replacement for BullMQ/Sidekiq with 10x throughput and 90% less memory. Backed by Redis. Python & Node.js bindings included.

dpdfnet

DPDFNet: causal single-channel speech enhancement that boosts DeepFilterNet2 with dual-path RNN blocks for stronger long-range temporal and cross-band modeling. Repo includes PyTorch implementation + checkpoints, ONNX & TFLite models with inference code, and a real-time demo.

conifer

BDT Inference for FPGAs

dbz-lib

Databento Binary Encoding (DBN) - Fast message encoding and storage format for market data

smelter-sdk

Python client for Smelter — a real-time, low-latency, programmable video and audio composition toolkit.

aiohttp-sendgrid

SendGrid mail send API wrapper

d2-sdk

Detect and Deny - Deterministic Function-Level Guardrails for AI Agents

sparsepixels

Efficient convolution for sparse data on FPGAs

disruptor-mp

Ultra-low-latency, high-throughput multiprocess transport over SHM and mmap. LMAX-Disruptor-style cross-process ring substrate.

myelon

Ultra-low-latency, high-throughput multiprocess transport over SHM and mmap. LMAX-Disruptor-style cross-process ring substrate.

residue-protocol

Bare-Metal AVX2 Inference Shield for LLMs

stimm

Dual-agent voice orchestration built on livekit-agents

python-socketio-rishavrmoney

Python Socket.IO server and client

hyperion-ai

Ultra-low-latency LLM gateway with microsecond caching, dynamic routing, budgets, analytics, and forecasting.

lltune

LLTune tunes Linux servers for ultra-low latency workloads. It scans system state, generates a config, audits for risks, applies safe CPU/ memory/network/kernel/IRQ tuning with backups and rollback, and stages persistence services for reboot.

securefabric-client

SecureFabric is a secure, low-latency messaging fabric designed for verified senders and end-to-end confidentiality.

hashlog-map

Ultra fast and low latency C++ Logging library with flexible configuration and high throughput in multithreaded environment

holohub

Central repository for Holoscan Reference Applications