> For the complete documentation index, see [llms.txt](https://ayakaleaf-pro.ayaka.space/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://ayakaleaf-pro.ayaka.space/on-premises/installation/introduction.md). # Introduction Deploy Ayakaleaf Pro with guidance on migration, requirements, and LaTeX performance. Ayakaleaf Pro is a drop-in replacement for the official Overleaf Community Edition, shipped as a Docker image. ## Choose your installation path | I want to… | Go to | | --------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------ | | Install on a new server | [Using Docker Compose on Linux](/on-premises/installation/using-docker-compose/linux.md) (starts with installing Docker) | | Run it locally on a Mac | [Using Docker Compose on macOS](/on-premises/installation/using-docker-compose/macos.md) (Docker Desktop) | | Let Claude or Codex do the install | [Using Claude or Codex](/on-premises/installation/using-claude-or-codex.md) | | Try it quickly in the browser (minimal TeX Live) | [Using GitHub Codespace](/on-premises/installation/using-the-github-codespace.md) | | Upgrade an existing official Overleaf Community Edition install | [Migrate from Existing](/on-premises/installation/migrate-from-existing.md) | | Install without internet access | [Air-gapped/offline deployments](/on-premises/installation/air-gapped-offline-deployments.md) | {% hint style="info" %} **Coming from the official Overleaf Server Pro?** ([What is the difference?](/on-premises/readme/ayakaleaf-pro-vs-server-pro.md)) Open an issue before migrating: templates are stored differently, and we'll help you for free.\ **Want to go back to the official Overleaf Community Edition?** Set `SERVER_PRO=false` (ayaka-notes Toolkit) or remove the override file (official Toolkit). No data is lost. {% endhint %} ### Hardware requirements We conducted relevant tests and found that, if you run Overleaf on a 2-core, 2 GB RAM server, its core components—primarily based on Node.js—will consume approximately 1 GB of memory. Redis and MongoDB may require an additional 0.5 GB, while starting a sandboxed Docker container for compilation can consume another 0.3 GB. This means that a 2-core, 2 GB server can realistically handle only about one concurrent compilation at a time. * The minimum hardware requirements for installing Ayakaleaf Pro are a 2-core CPU and 3 GB of RAM. * Due to the size of our TeX Live image, at least 40 GB of storage is required to enable sandboxed compilation. * If you want to install the complete set of TeX Live images from 2020 through 2026, please make sure you have at least 150 GB of free disk space available. Here are some recommended hardware configurations for different usage scenarios: | Use Case | Recommended Configuration | | ---------------------------------------------------- | ---------------------------------------------------------------------------------------------------- | | Personal use, with occasional collaboration | 2 CPU cores, 4 GB RAM, 80 GB storage | | Small teams, up to 10 users with light collaboration | 2 CPU cores, 8 GB RAM, 256 GB storage | | Medium-sized teams, up to 100 users | 4 CPU cores, 16 GB RAM, 1 TB storage | | Large-scale or university deployments | Distributed cluster deployment, with S3 storage; please contact us if you need deployment assistance | ### Software requirements For software requirements, we recommend using a recent and stable version of Ubuntu (e.g. Ubuntu 24.04 in 2025–2026, Ubuntu 22.04 in 2023–2024) together with the latest stable version of Docker. {% hint style="info" %} Ubuntu 26.04 ships with Linux kernel 7.0, and MongoDB refuses to start on Linux kernel 6.19 through 7.0.13 ([details](https://www.mongodb.com/docs/manual/administration/production-notes/)). To run on these kernels, add the following to the `mongo` service in `overleaf-toolkit/config/docker-compose.override.yml` (the same file that sets the Ayakaleaf Pro image). Our CI runs on Ubuntu 26.04 with this setting. Kernel 7.0.14 and later don't need it. {% code title="docker-compose.override.yml" %} ```yml services: mongo: environment: GLIBC_TUNABLES: glibc.pthread.rseq=1 ``` {% endcode %} {% endhint %} We do not recommend deploying Ayakaleaf Pro on Windows or other unsupported operating systems like CentOS, as you may encounter compatibility issues or unexpected problems during installation and operation. But if you would like to try deploying Ayakaleaf Pro on ARM-based systems or macOS, we welcome you to share your experience and feedback with the community. ### LaTeX Performance LaTeX is a single threaded program, meaning it can only utilize one CPU core at a time. When compiling a single document, performance is primarily limited by the CPU's single-core clock speed. Therefore the faster the single core performance of your CPU the faster you will be able to compile a document. More CPU cores are only beneficial when multiple users are compiling documents concurrently. Using the [ThuThesis template](https://www.overleaf.com/latex/templates/thuthesis-tsinghua-university-thesis-latex-template/cfwgcxtvkbsx) as an example, the compilation time is about 6 seconds on an Intel Core i9-14900K (5.6 GHz), while it takes about 13 seconds on an Azure Standard D2as v5 instance (2 vCPUs, 8 GiB RAM) powered by an AMD EPYC 7763 (3.2 GHz). Of course, actual compilation performance depends on your hardware configuration. Most server CPUs do not run at such high clock speeds as 14900K, so you will need to make a trade-off between cost and compilation speed. Meanwhile, we have also conducted an in-depth analysis of LaTeX compilation performance. If you are interested, please refer to our blog [Overleaf Benchmark: A Deep Research of Concurrent LaTeX Compilation in Overleaf (SaaS and Self-host)](/blog/2026/overleaf-benchmark.md). --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://ayakaleaf-pro.ayaka.space/on-premises/installation/introduction.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.