Managed compute
Choose CPU or GPU resources for an endpoint. Review the price before starting and configure an idle timeout.
Managed cloud endpoints
Explore endpointsEdelux Foundry
Build agents
with your team.
Configure tools and knowledge, review agent runs, and work on projects together. Keep access, budgets and run history in your workspace.
Business workspace billing is separate from Personal and API credit. Review payment availability in your account.
One workspace. Four starting points.
Prompt + knowledge
Agent + tools
Run history
Give an agent its prompt, its tools and its knowledge. Inspect every run before it reaches a workflow.
Build an agentThe business platform
One workspace,
for your team.
Build agents, connect your tools, and review what ran. Business features depend on your organization’s permissions and configured services.
Go from a prompt to a reviewed, versioned, running system.
- Chat
Compare answers across models, search your own knowledge, and build a small app from the conversation.
- Projects & IDE
A browser editor with a terminal and checkpoints, where a coding agent works on your repository under review.
- Agents
Give an agent its prompt, its tools and its knowledge, then inspect every run it produces.
- Workflows
Chain agents into asynchronous pipelines, with retries and an approval step where a human decides.
- Blueprints
Package an agent or workflow as a portable version you can review, publish and roll back.
- Prompt Library
Keep reviewed prompts in one place so teams reuse what already works.
Model registry
Choose the model.
Keep your options open.
Open-weight releases, connected to the serving engine your infrastructure can actually run.
vLLMSGLangTGIOllama
55 matching models
Moonshot AI
Kimi K3
2.8T MoE, 104B active, 1M context
Kimi K3 License
Alibaba Qwen
Qwen3.8 Max
2.4T MoE, 95B active, 262K context
Qwen3.8-Max License
DeepSeek
DeepSeek V4 Pro
1.6T MoE, 49B active, 1M context
MIT
Z.ai
GLM-5.2
753B MoE, 40B active, 1M context
MIT
MiniMax
MiniMax-M3
428B MoE, 23B active, 1M context
MiniMax Community License
DeepSeek
DeepSeek V4 Flash
284B MoE, 13B active, 1M context
MIT
Model links open upstream weights. Inclusion in this roster does not mean a hosted endpoint is running.
Run where your workload belongs.
Use managed compute for experiments, or take a deployment bundle to your own infrastructure. Keep the deployment artifact either way.
Your infrastructure
Download a portable bundle, review its configuration, and apply it to a Kubernetes cluster, Compose host or Linux server.
Kubernetes · Docker · Linux
Explore bundlesKeep the tools you already use.
The setup script detects the coding tools already installed on your machine and writes the gateway endpoint, token and model profiles into each one.
- 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
Give teams access.
Keep the record.
Control who can run workloads, what they may spend, where they may connect, and which steps need a human decision.
Scope every surface
Roles and scoped API keys decide who reaches which workspace, checked on each request rather than in the interface.
Cap the spend
Budgets, quotas and per-team chargebacks, with alerts that fire before a limit is passed.
Keep a human in the loop
Require an approval inside a workflow, assign the reviewer, and store the decision with the run.
Detect tampering
Consequential actions append to a per-organisation SHA-256 hash chain that a verification endpoint can replay.
Classify under the AI Act
Register a system, get its risk tier and obligations, and track the reporting windows the regulation sets.
Fence the network
An egress policy restricts the domains a workload may reach, in audit mode first, then enforced.
Capacity market
GPUs are an asset.
Treat them like one.
Most clusters are either saturated or half empty, rarely correctly sized. Foundry puts both sides in the same book, and settles the result in your ledger.
Buy what you lack
Discover MCP servers, blueprints and compute packs published by other organisations, with the contract terms attached.
Browse the marketplaceSell what sits idle
List spare capacity on the order book. Matching, commission and settlement are recorded against your wallet.
Open the exchangeServe your own clients
Resell inference behind per-client keys, or sign take-or-pay wholesale commitments with SLA probing.
See resaleBring your own hardware.
Sign in to build an agent, deploy a model, connect a cluster, and see real usage and cost per team.