Buy what you lack
Discover MCP servers, blueprints and compute packs published by other organisations, with the contract terms attached to each listing.
The deployment platform for European AI
Deploy AI in Europe.Run it where you choose.
Edelux brings the model gateway, the workspace, and the deployment path into one console. Start on managed compute, use capacity listed by partners and companies, or run on infrastructure you control.
- product surfaces
- 28
- models listed
- 55
- verified entries
- 14
Deploy anatomyIllustrative example
- ChooseDeepSeek V4 Pro
deepseek-ai/DeepSeek-V4-Pro · MIT - TargetManaged compute, listed capacity, or your own cluster.
- CallOne gateway behind scoped API keys, with the request history beside the work.
Illustrative anatomy only. No request is sent, and nothing here runs or bills.
From first prompt to a running endpoint.
Chat across models, review the agent's work, and ship from one workspace. Connect the tools you already use through one gateway and scoped keys.
- ChooseOpen a model from the real Edelux registry.
- WorkChat, edit in the IDE, and inspect each agent run.
- ShipStart an endpoint, or call it through a scoped API key.
Live marketplace proof
Europe runs its spare compute in one book.
Partners and companies list the GPU capacity they spare, and teams buy what they lack. Matching, commission and settlement are recorded in your own ledger, so every position stays inspectable.
- 55
- 4
- 28
- Browse the marketplace
- List spare capacity
Contribute what sits idle
List spare GPU capacity from your own fleet. Matching, commission and settlement are recorded against your own wallet, so the position is inspectable.
- See resale
Serve your own clients
Resell inference behind per-client keys, or sign take-or-pay wholesale commitments whose endpoints are probed against the contract.
Deploy path
Deploy on shared capacity, or run it at home.
Start on hosted compute to try something out, then move it to your own datacentre or private cloud. The deployment artifact is the same either way.
Hosted compute
Choose CPU or GPU resources for a private endpoint. Review the price before you start, and set an idle timeout so an experiment does not bill all night.
Your own datacentre or private cloud
Download a portable bundle, read its configuration, and apply it to a Kubernetes cluster, a Compose host, a bare Linux server or your own private cloud. Nothing is hidden in a vendor console.
- Kubernetes
- Docker
- Google Cloud
- Microsoft Azure
- AWS
- OVHcloud
- Scaleway
- Bare metal Linux
The platform
Everything the work needs, in one place.
Pick an area to see what it does. Each one is a real screen in the product, not a capability claim.
Go from a prompt to a reviewed, versioned, running system.
- ChatCompare answers across models, search your own knowledge, and build a small app from the conversation.
- Projects & IDEA browser editor with a terminal and checkpoints, where a coding agent works on your repository under review.
- AgentsGive an agent its prompt, its tools and its knowledge, then inspect every run it produces.
- WorkflowsChain agents into asynchronous pipelines, with retries and an approval step where a human decides.
- BlueprintsPackage an agent or workflow as a portable version you can review, publish and roll back.
- Prompt LibraryKeep reviewed prompts in one place so teams reuse what already works.
Models for shared capacity
Open-weight models Europe can run anywhere.
55 releases, each importable through the same path and servable on shared capacity. Curated entries carry their verified size and context.
- vLLM
- SGLang
- TGI
- Ollama
55 models
- Kimi K3Moonshot AI2.8T MoE, 104B active, 1M contextKimi K3 LicenseCuratedOpen the upstream weights on Hugging Face
- Qwen3.8 MaxAlibaba Qwen2.4T MoE, 95B active, 262K contextQwen3.8-Max LicenseCuratedOpen the upstream weights on Hugging Face
- DeepSeek V4 ProDeepSeek1.6T MoE, 49B active, 1M contextMITCuratedOpen the upstream weights on Hugging Face
- GLM-5.2Z.ai753B MoE, 40B active, 1M contextMITCuratedOpen the upstream weights on Hugging Face
- MiniMax-M3MiniMax428B MoE, 23B active, 1M contextMiniMax Community LicenseCuratedOpen the upstream weights on Hugging Face
- DeepSeek V4 FlashDeepSeek284B MoE, 13B active, 1M contextMITCuratedOpen the upstream weights on Hugging Face
Model links open the upstream weights on Hugging Face. Being listed here does not mean a hosted endpoint is already running.
Workspace
One workspace from first prompt to production.
Chat, projects, agents, runs, endpoints and costs stay in the same place, with one gateway and scoped keys connecting the tools you already use.
Edelux workspaceIllustrative preview
From idea to inference
One connected workspace.
Build
Chat, agents & tools
Connect
One model gateway
Control
Access, budgets & audit
Example requestPOST
/resale/v1/chat/completionsDeepSeek V4 ProReview an incident report
Scoped API keysRequest historyYour deployment target
From first prompt to a running endpoint.
Chat across models, review the agent's work, and ship from one workspace. Connect the tools you already use through one gateway and scoped keys.
- ChooseOpen a model from the real Edelux registry.
- WorkChat, edit in the IDE, and inspect each agent run.
- ShipStart an endpoint, or call it through a scoped API key.
Workspace and developers
One gateway, one set of keys, one usage record.
The setup script finds the coding tools installed on your machine and writes the gateway endpoint, token and model profiles into each one. Scoped keys and metered usage follow the work.
- 01
Read how a run is put together
Every agent run keeps its trace, its token usage and its outcome, and you can read the whole thing before it reaches anyone else.
See a run - 02
Point your editor at the gateway
One command writes the gateway endpoint, token and model profiles into the coding tools already on your machine. Scoped keys decide what each tool may reach, and usage is metered per key.
Issue a key - 03
Decide where the model runs
Bring a model you already run, or let Edelux start one on managed compute. Either way the gateway endpoint keeps the same shape.
Start an endpoint
- Oh My Pi
~/.omp/agent/models.yml - Claude Code
~/.claude.json - Codex CLI
~/.codex/config.json - OpenCode
~/.config/opencode/opencode.json - Edecode
~/.config/edecode/edecode.json - Hermes Agent
~/.hermes/config.yaml - Continue.dev
~/.continue/config.json - Aider
~/.aider.conf.yml
Sovereignty evidence
Keep the workloads, the spend, and the record at home.
Control who can run workloads, what they may spend, where they may connect, and which steps need a human decision. Each control below opens the screen that enforces it.
Scope every surface
Roles and scoped API keys decide who reaches which workspace, checked on each request rather than in the interface.
Manage team & rolesCap the spend
Budgets, quotas and per-team chargebacks, with alerts that fire before a limit is passed.
Set budgetsKeep a human in the loop
Require an approval inside a workflow, assign the reviewer, and store the decision with the run.
Add an approvalDetect tampering
Consequential actions append to a per-organisation SHA-256 hash chain that a verification endpoint can replay.
Open the audit trailTrack AI Act obligations
Register a system, see its risk tier and obligations, and track the reporting windows the regulation sets. The record supports your process; it does not certify it.
Register a systemFence the network
An egress policy restricts the domains a workload may reach, in audit mode first, then enforced.
Configure egress
Start here
Open a workspace, or connect your editor first.
Signing in gives you the assistant, the agents, the endpoints and the cost record. Connecting an editor first is the smaller step, and it needs nothing but a key.