Your hardware · your models · your data
Sovereign AI
starts at your desk
A complete AI development environment on your own machines: notebooks, your GPUs and your models, and a cluster to try what you build, all driven from a coding agent. Models, components and services load and unload as you need them, so your machine's limited resources go to the work you are doing now.
- Get models
- Fine-tune
- Build apps
- Evaluate
Every task is a sentence to your agent
Thinkube is designed for you to work from a coding agent. The platform gives the agent its own tools for every part of it: when you ask, it sees which models are loaded, what each GPU is holding, how a deploy is going and what a service logged, and it runs the same operations you would run by hand. You decide what happens: the agent changes something only when you ask. Every playbook on this site starts with the sentence to say.
Four areas, one platform
The base, ready and signed in
Unmodified upstream Kubernetes on your machines, with identity, storage, registry, database and GPUs ready, driven from a coding agent.
Thinkube GitOpsPush to git, and it runs
From a template to an application at its own address behind your login. Each later change goes live when you push.
Thinkube ModelsServe, fine-tune, track
Open models on your own GPUs. Use them from notebooks and from your apps, at one address.
Thinkube TandemYou and an agent, together
Ask in a chat, or build a change step by step. The agent does the work; you decide what happens.
First time here?
Build a web app from the template
beginnerA full-stack app with a database and a login, at its own address, from one sentence
2026-10-04 Thinkube Models 30 minServe your first model
beginnerLoad an open model on your own GPU and call it from a notebook, as you would call a cloud API
2026-10-04 Thinkube Tandem 1 hBuild your first application change with Tandem
beginnerWrite what you want in plain sentences, see how they were read, sign the promises, and get the change running in your app
2026-10-04All playbooks
Check and restart services from your agent
beginnerAsk what is running and whether it is healthy, restart a service, and turn one off and on again, from a chat with your agent
2026-10-04 Thinkube Kubernetes 30 minInstall and remove optional components
beginnerAdd a database, a vector store or an inference backend to the platform when you need it, and take it out again
2026-10-04 Thinkube GitOps 30 minBuild a web app from the template
beginnerA full-stack app with a database and a login, at its own address, from one sentence
2026-10-04 Thinkube GitOps 30 minChange your app and ship it
beginnerEdit the code, run its tests, push, and the new version is live
2026-10-04 Thinkube GitOps 30 minPass configuration to your app
beginnerValues your app reads at run time, and the few the browser may see
2026-10-04 Thinkube GitOps 30 minPublish your app as a template
beginnerTurn an app you improved into a template anyone on your cluster can deploy
2026-10-04 Thinkube GitOps 30 minRun a pipeline from your app
intermediateMulti-step jobs your app submits and Argo Workflows runs, each step in your app's own image
2026-10-04 Thinkube GitOps 30 minDeploy a serverless service
beginnerA service that runs only while it is being called, like AWS Lambda, on your own machines
2026-10-04 Thinkube GitOps 30 minStore and fetch files with the file gateway
beginnerUpload, list, download and delete files in Thinkube Storage over a REST API and a web page
2026-10-04 Thinkube GitOps 30 minConvert papers to Markdown and JATS XML
intermediateTurn PDF papers into structured text with Docling, on CPU in a pipeline step or with a vision model on your GPU
2026-10-04 Thinkube Models 30 minServe your first model
beginnerLoad an open model on your own GPU and call it from a notebook, as you would call a cloud API
2026-10-04 Thinkube Models 30 minLoad a model on the node and context you choose
intermediatePick the GPU node, the backend and the context length for a load, and what happens when a node is busy
2026-10-04 Thinkube Models 30 minCall your models from an app or a script
beginnerPoint the OpenAI or Anthropic client you already use at your gateway, with a Thinkube API token
2026-10-04 Thinkube Models 30 minBuild a notebook environment
beginnerStart from agent-dev or fine-tuning, add the packages you need, and get a kernel built for every architecture in your cluster
2026-10-04 Thinkube Models 30 minRun notebooks from your agent, watched or unattended
beginnerHave your agent run a notebook while you watch it in a Thinkube IDE tab, or on a server of its own that stops when the run ends
2026-10-04 Thinkube Models 30 minBuild a research assistant over your papers
intermediateIndex arXiv papers in your own vector store, ask questions that come back with sources, and let agents argue from the index
2026-10-04 Thinkube Models 9 h 30 minFine-tune a model on rewards a program checks
advancedTrain a 4B model with GRPO against a solver that grades every answer, with no labelled data, and register the result beside its run
2026-10-04 Thinkube Models 30 minServe a model you fine-tuned
intermediateGive a registered fine-tune a catalogue entry, load it on a GPU and call it through the gateway like any other model
2026-10-04 Thinkube Tandem 2 hServe the model Thinkube Tandem Chat uses
beginnerMirror and load the model behind the gateway alias thinkube-fast, then ask your first question in a Thinkube Tandem Chat
2026-10-04 Thinkube Tandem 1 hBuild your first application change with Tandem
beginnerWrite what you want in plain sentences, see how they were read, sign the promises, and get the change running in your app
2026-10-04 Thinkube Tandem 30 minBuild the improvements Tandem’s reviewers suggest
beginnerKeep the findings a report lists, put them in the box as sentences, and build them without touching what is already built
2026-10-04