[{"data":1,"prerenderedAt":114},["ShallowReactive",2],{"seo-verification":3,"marketplace-app-en-mlflow":6},{"google":4,"bing":5},"EycwPY2XMyTkVzas3n1ygeNJFGAH513qrMjfDljzsMQ","",{"key":7,"data":8},"marketplace-app-en-mlflow",{"slug":9,"slugs":10,"categorySlugs":11,"name":16,"description":17,"phase":18,"unavailableReason":19,"docsUrl":20,"logo":21,"github":22,"tagline":23,"longDescription":24,"features":25,"useCases":32,"steps":42,"faq":58,"specs":71,"compatibleOs":83,"relatedApps":84,"relatedPosts":107,"category":111},"mlflow",{"fr":9,"en":9,"ar":9,"es":9},{"fr":12,"en":13,"ar":14,"es":15},"ia","artificial-intelligence","الذكاء-الاصطناعي","inteligencia-artificial","MLflow","Machine learning experiment tracking, model registry and LLM application tracing, on your VPS and behind a password.",2,"awaiting_qa","https:\u002F\u002Fservorbit.com\u002Fblog\u002Fself-host-mlflow-vps-experiment-tracking","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F39938107","https:\u002F\u002Fgithub.com\u002Fmlflow\u002Fmlflow","MLflow — track your ML experiments, version your models and trace your LLM apps on your VPS, over HTTPS and protected.","MLflow is the reference open-source platform (Apache-2.0 licensed) for the machine learning lifecycle. A central tracking server receives, from your notebooks and training scripts, the parameters, metrics and artifacts of every run: you compare your attempts side by side instead of copying them into a spreadsheet.\n\nThe model registry versions what goes to production, with aliases and history. And for generative AI applications, MLflow traces every call to a language model — inputs, outputs, latency, intermediate steps — so you can understand an answer and debug it. On ServOrbit, the server installs in one click, protected by an administrator account and served over HTTPS.",[26,27,28,29,30,31],"Experiment tracking — parameters, metrics and artifacts of every run, compared in the interface.","Model registry — versions, aliases and history of the models headed for production.","LLM tracing — inputs, outputs, latency and steps of your generative AI applications.","Artifacts served by the server — files, charts and models uploaded and downloaded over HTTPS, no S3 storage to configure.","Every Python library — scikit-learn, PyTorch, Transformers, XGBoost, OpenAI, LangChain, with automatic logging.","Protected access — admin account with a generated password, required on the interface and the API; other accounts only see what is shared with them.",[33,36,39],{"title":34,"body":35},"Compare your training runs","Every notebook or script run records its hyperparameters and metrics: you find your top-scoring attempt on one screen, with its code and files.",{"title":37,"body":38},"A model registry for the team","Register validated models, give them an alias such as \"production\", and let your inference service always load the right version.",{"title":40,"body":41},"Debug an LLM application","Turn on tracing for your assistant or agent: every request shows its model calls, its steps and its latency, so you can understand a bad answer.",[43,46,49,52,55],{"title":44,"body":45},"Create the VPS and choose the address","Order an Ubuntu 24.04 VPS with 2 GB of RAM (4 GB and more disk if you store many models). MLflow requires a domain: your scripts, notebooks and team connect to it, so it must answer over HTTPS on a stable name. It opens on a subdomain of one of your domains, or on the VPS's free subdomain (mlflow.{dns_slug}.servorbit-dns.com), HTTPS included. No need to buy a domain name.",{"title":47,"body":48},"Install MLflow from 1-click apps","In your client area, select MLflow, choose the address and click Install. The script deploys the official container, generates the `admin` account's password, enables authentication on the interface and the API, limits the accepted host names to your address and configures the nginx reverse proxy over HTTPS.",{"title":50,"body":51},"Log in for the first time","Open https:\u002F\u002Fyour-domain\u002F: the browser asks for credentials. Enter `admin` and the password shown in your client area. You can change it later through the `users\u002Fupdate-password` API; the new password is kept across restarts.",{"title":53,"body":54},"Connect your notebooks and scripts","Set `MLFLOW_TRACKING_URI=https:\u002F\u002Fyour-domain`, `MLFLOW_TRACKING_USERNAME=admin` and `MLFLOW_TRACKING_PASSWORD`, then call `mlflow.start_run()` in your code. Parameters, metrics and artifacts appear in the interface right away.",{"title":56,"body":57},"Open access to your team","Create one account per person with the `users\u002Fcreate` API, logged in as `admin`. A new account sees no experiment until you share one with it: permissions are granted experiment by experiment and model by model.",[59,62,65,68],{"q":60,"a":61},"What is the difference between MLflow and Langfuse?","Langfuse focuses on LLM applications: traces, evaluations and prompt management. MLflow covers the whole machine learning lifecycle — training tracking, run comparison, model registry — and also traces generative AI applications. For a team that trains its own models, MLflow is the right starting point; for an application that only calls LLMs, Langfuse is more specialised.",{"q":63,"a":64},"Do I need a domain name for MLflow?","Yes. Your scripts, notebooks and team connect to the server, and MLflow only accepts declared host names: it must answer over HTTPS on a stable name. No need to buy one: every ServOrbit VPS comes with a free subdomain (mlflow.{dns_slug}.servorbit-dns.com) with an HTTPS certificate. You can also use a subdomain of one of your domains.",{"q":66,"a":67},"How do I log in for the first time?","Open https:\u002F\u002Fyour-domain\u002F and enter the `admin` user and the password shown in your client area. These credentials also protect the API: your code reads them from `MLFLOW_TRACKING_USERNAME` and `MLFLOW_TRACKING_PASSWORD`.",{"q":69,"a":70},"Where are my models and artifacts stored?","On your VPS, in the `mlflow-data` Docker volume: the SQLite experiment database, the accounts database and the artifacts folder. Nothing goes to a third party. To back up, stop the container and archive this volume, then copy the archive off the VPS.",{"ram":72,"cpu":73,"disk":74,"stack":75,"port":79,"license":80,"version":81,"os":82},"2 GB minimum","1 vCPU","20 GB",[76,77,16,78],"Docker","Python","SQLite","8111","Apache-2.0","v3.18.0","Ubuntu 24.04",[],[85,93,101],{"name":86,"slug":87,"categorySlug":13,"categoryName":88,"categoryColor":89,"logo":90,"tagline":91,"description":92},"Langfuse","langfuse","Artificial Intelligence","text-purple-400 bg-purple-500\u002F10","https:\u002F\u002Fcdn.jsdelivr.net\u002Fgh\u002Fselfhst\u002Ficons\u002Fsvg\u002Flangfuse.svg","Trace, evaluate and manage your LLM applications — self-hosted, production-ready.","Open-source LLM observability: trace every AI request, run evaluations, manage prompt versions and compare model costs — self-hosted, compatible with OpenAI, LiteLLM, LangChain and Ollama.",{"name":94,"slug":94,"categorySlug":95,"categoryName":96,"categoryColor":97,"logo":98,"tagline":99,"description":100},"marimo","development","Development","text-warning bg-warning\u002F10","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F98563464","marimo — reactive Python notebooks on your own VPS: Git-friendly .py files, cells that recompute themselves, data and charts.","Reactive Python notebook, open source: notebooks stored as plain .py files, cells that recompute automatically, data libraries included.",{"name":102,"slug":103,"categorySlug":13,"categoryName":88,"categoryColor":89,"logo":104,"tagline":105,"description":106},"Ollama","ollama","https:\u002F\u002Fcdn.simpleicons.org\u002Follama","Run open-source LLMs on your VPS — OpenAI-compatible API, zero per-token cost.","Run open-source LLMs locally via a dead-simple API. Pull Llama 3, Mistral, Qwen or DeepSeek in one command — OpenAI-compatible, zero per-token cost.",[108,109,110],"self-host-langfuse-vps","self-host-marimo-vps","heberger-ollama-vps",{"key":12,"slug":13,"name":88,"objective":112,"icon":113,"color":89},"Build, host and run AI solutions.","ai",1793982646993]