Godzilla Community
godzilla.dev is an open-source C++/Python infrastructure for self-hosted crypto funding rate arbitrage and market making, with ultra low-latency and enterprise private deployment
Install / Use
npx skills add godzilla-foundation/godzilla-communityInstalls into whichever agent you are using.
README
godzilla.dev
godzilla.dev is an open-source C++/Python infrastructure for self-hosted crypto funding rate arbitrage and market making, with ultra low-latency architecture and enterprise private deployment.
Documentation · Whitepaper · Installation · FAQ · Enterprise
</div><p align="center"> <img src="./scripts/benchmark/figures/fig2_tick_to_trade_cdf.png" alt="Tick-to-trade latency benchmark CDF" width="860" /> </p> <p align="center"><strong>Median tick-to-trade latency is ~125 μs.</strong></p> <p align="center"><em>Tick-to-trade latency benchmark from the reproducible single-host artifact.</em></p>
What is godzilla.dev?
godzilla.dev is not a retail trading bot — it is production-grade trading infrastructure. It combines an ultra low-latency C++ execution core with a flexible Python strategy layer, and is designed to run co-located next to exchange matching engines (e.g. AWS availability zones used by major exchanges).
Trading teams and individual quants use godzilla.dev to run:
- Funding rate arbitrage — delta-neutral spot–perp and cross-exchange funding rate strategies on centralized exchanges
- Market making — ultra low-latency pure market making that quotes both sides of the book
- Custom low-latency strategies — any automated strategy built on the same execution core
The codebase is free and publicly available under the Apache 2.0 license. Our mission is to democratize high-frequency trading: infrastructure that was once exclusive to quant funds, now open to everyone.
Why godzilla.dev?
- C++ execution core. Order placement, cancellation, and market data handling run in C++ for microsecond-level tick-to-trade latency — while strategies stay in Python for fast iteration. In the reproducible journal-only benchmark (WHITEPAPER.md), the native path shows a median tick-to-trade of ~125 μs measured on a commodity laptop-class CPU; co-located server hardware with tuned kernels represents a different operating point.
- Self-hosted and private. godzilla.dev is local software, not a web platform. Your API keys, private keys, and strategy configuration never leave machines you control.
- Co-location ready. Installed from source and managed with
pm2, following the same workflow you use to deploy on a co-located AWS machine. - Open source, community-driven. Exchange connectors and strategy templates are maintained by the Godzilla Foundation together with a global community of algo traders.
- Enterprise private deployment. For funds and funding rate arbitrage teams that need a fully self-hosted stack with support, see godzilla.dev Enterprise.
How is godzilla.dev different from typical bot frameworks?
| | godzilla.dev | Typical Python bot frameworks | |---|---|---| | Execution core | C++ (Python strategy layer) | Pure Python | | Latency profile | Designed for co-located, microsecond-level HFT | Second/millisecond-level, latency-tolerant strategies | | Deployment | Self-hosted, co-location first | Local or cloud, latency not a design goal | | Primary use cases | Funding rate arbitrage, HFT market making | Grid/DCA bots, general automation | | Target users | Prop traders, market makers, arbitrage teams | Retail users, hobbyists |
Frameworks like Hummingbot are excellent for breadth of connectors and community strategies; godzilla.dev focuses on latency-critical, self-hosted execution for professional market making and arbitrage.
Getting started
- Read the Installation guide — we recommend installing from source, the same way you deploy to a co-located machine.
- Explore the Strategies documentation and run your first strategy.
- Follow the Learning series — step-by-step tutorials from Python basics to AI-assisted quant workflows.
Community
- Telegram community — connect with other algo traders and get help
- YouTube @godzilla-dev — tutorials and walkthroughs
- FAQ — answers to common questions
All official Godzilla Foundation code lives in repositories under this GitHub organization. Please download godzilla.dev software only from these official sources.
See BRAND.md for official descriptions and press assets
License
Apache 2.0 — free for personal, research, and commercial use, including forks and private modifications.
<div align="center">
中文简介
</div>godzilla.dev 是一个开源的 C++/Python 基础设施,用于自托管(self-hosted)的加密货币资金费率套利与做市,具备超低延迟架构,并支持企业级私有化部署。
godzilla.dev 不是零售型交易机器人,而是生产级的交易基础设施:C++ 撮合级低延迟执行内核 + Python 策略层,为交易所同机房(co-location)部署而设计。交易团队和个人量化交易者用它来运行:
- 资金费率套利 —— 现货–永续、跨交易所的 Delta 中性资金费策略
- 做市 —— 超低延迟的双边挂单做市
- 自定义低延迟策略 —— 基于同一执行内核构建任意自动化策略
代码库基于 Apache 2.0 开源许可证免费公开。我们的使命是让高频交易民主化:把曾经只属于量化基金的基础设施,开放给每一个人。
企业级私有化部署(面向资金费率套利团队与机构)请见 godzilla.dev Enterprise。
Citation
If you use godzilla.dev or the benchmark artifact in your research, please cite:
Kun Xue. Low-Latency Execution Infrastructure for Funding-Rate Arbitrage: Architecture and a Reproducible Single-Host Benchmark. 2026. Benchmark artifact: https://doi.org/10.5281/zenodo.21307411
Related Skills
dbx
13.7k20 MB lightweight cross-platform database client for 70+ databases, including MySQL, PostgreSQL, SQLite, Redis, MongoDB, DuckDB, SQL Server, and Dameng. Built-in AI, MCP Server, CLI, desktop and Docker.
product
Cloud-agnostic Kubernetes infrastructure with Terraform & Helm for homelabs, edge, and production clusters.
ghidra-mcp
3.2kGhidra MCP Server — 200+ MCP tools for AI-powered reverse engineering. GUI plugin + headless server, lazy tool loading, convention enforcement, batch operations, Ghidra Server integration, and Docker deployment.
radar
2.8kThe missing open-source Kubernetes UI with a built-in MCP server for AI agents. See what's broken, why, and what changed. Issues, Topology, event timeline, Helm, GitOps, live service traffic, and cluster audits - all in one Go binary.
