Median tick-to-trade latency is ~125 μs.
Tick-to-trade latency benchmark from the reproducible single-host artifact.
## 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](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](https://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 1. Read the [Installation guide](https://godzilla.dev/documentation/installation/) — we recommend installing from source, the same way you deploy to a co-located machine. 2. Explore the [Strategies documentation](https://godzilla.dev/documentation/strategies/) and run your first strategy. 3. Follow the [Learning series](https://godzilla.dev/learning/) — step-by-step tutorials from Python basics to AI-assisted quant workflows. ## Community - [Telegram community](https://godzilla.dev/documentation/community/) — connect with other algo traders and get help - [YouTube @godzilla-dev](https://www.youtube.com/@godzilla-dev) — tutorials and walkthroughs - [FAQ](https://godzilla.dev/documentation/faq/) — answers to common questions All official Godzilla Foundation code lives in repositories under [this GitHub organization](https://github.com/godzilla-foundation). Please download godzilla.dev software only from these official sources. See [BRAND.md](./BRAND.md) for official descriptions and press assets ## License Apache 2.0 — free for personal, research, and commercial use, including forks and private modifications. ---