Open source · Apache 2.0 · Self-hosted

The AI infrastructure that turns prompts into optimal workflows.

MoE Sovereign analyses each request, combines the right models, Expert Templates, GraphRAG, MCP tools and current research—then validates the result and retains useful knowledge for future tasks.

Not just a router. A compiler for executable AI workflows.

Explore the architectureTechnical documentation
Peer-reviewed Resource AwardEuroHPC JU · LUMI-G4,500 node-h

One API, many capabilities

One LLM to clients.
A full system inside.

OpenAI Chat CompletionsOpenAI ResponsesAnthropic MessagesOllama API
POST /v1/responses
{
  "model": "moe-sovereign",
  "input": "Assess the current situation …"
}

Whitepaper · Preview Release

Architecture, research and evidence—open for review.

All three documents are preview editions. Content, methodology, results and pagination may change before final publication. Feedback is explicitly welcome.

Preview · Juli 2026

German Whitepaper

Complete German-language presentation of the platform, architecture and research vision.

Length
118 pages
Format
PDF · 3.1 MB
Download PDF ↓
Preview · July 2026

English Whitepaper

Complete international edition covering the platform, architecture and research vision.

Length
108 pages
Format
PDF · 3.1 MB
Download PDF ↓
Preprint · Preview

Scientific short paper

Concise English research edition for scientific discussion and review.

Length
12 pages
Format
PDF · 317 KB
Download preprint ↓

Publication status: Preview / preprint. Not yet released as a final edition and—unless stated otherwise in the document—not peer-reviewed.

Same prompt. Different process.

A model answers. MoE Sovereign organises the solution.

The architecture spends compute where freshness, knowledge, multiple steps or evidence justify it.

Direct model operation

Native LLM

  1. PromptOne input
  2. One modelStatic training knowledge
  3. AnswerNo persistent learning experience

For simple, stable tasks, this direct path may be optimal.

Adaptive processing

MoE Sovereign

  1. Prompt analysisDomain, complexity, risk
  2. Expert TemplateModels, dependencies, policies
  3. ExecutionGraphRAG · MCP · research · agents
  4. ValidationJudge · sources · correction loop
  5. Answer + knowledgeUseful results persist

The greatest advantage emerges for complex, current, knowledge-intensive or safety-relevant tasks.

The learning architecture

Knowledge grows outside model weights.

Today, a large model solves a novel task. Tomorrow, a smaller model can reuse verified knowledge and a proven workflow.

01ProcessExecute workflow
→
02ValidateCheck quality
→
03ExtractFacts, sources, paths
→
04PersistGraph + memory
→
05ImproveRouting + next request
Implemented

Production system learning

GraphRAG, layered caches, episodic knowledge, Correction Memory, feedback and routing signals improve system context—without uncontrolled training.

In development

Extend SLM competence

Successful plans and verified decisions are intended for distillation into specialised small models that can take over recurring work locally.

Research goal

Dynamic template generation

A LUMI-G-developed MoE Sovereign is intended to generate an optimal execution plan for each request, including resources, rules and persistence decisions.

One sovereign ecosystem

Three projects. Three clearly separated responsibilities.

MoE Sovereign remains the independently deployable core. Libris and Codex optionally extend it with federated knowledge exchange and an auditable data platform.

S
Core · Orchestration

MoE Sovereign

Compiles prompts into workflows of models, Expert Templates, knowledge, tools and validation. Operates fully without its sibling projects.

  • Adaptive AI workflows
  • GraphRAG, MCP and memory
  • OpenAI, Anthropic and Ollama APIs
Core repository ↗
L
Federation · Knowledge

MoE Libris

Federated knowledge exchange between sovereign instances—voluntary, trust-based and reviewed before acceptance.

  • JSON-LD push and pull
  • Pre-audit and human review
  • mTLS and sender-side privacy scrubbing
moe-libris.org ↗
C
Data · Audit

MoE Codex

Optional sovereign data and audit platform for organisations requiring catalogues, approvals, lineage and evidence.

  • Data catalogue and versioning
  • Approvals, lineage and drift detection
  • Policies, search and analytics tools
moe-codex.org ↗

Modular, not monolithic: Each project has its own responsibility and can be developed or operated independently. Integrations use documented interfaces.

Quality through interaction

Not more features. Better causal chains.

01

Better knowledge

GraphRAG, documents and current research provide grounded context beyond training knowledge.

02

Precise execution

MCP precision tools and structured tool paths execute deterministic tasks reliably.

03

Better decisions

Expert Templates, complexity analysis and routing select only necessary resources.

04

Higher reliability

Judge, Self-Correction, contradiction detection and source ranking check before output.

05

Continuous improvement

Corrections, feedback, episodes and routing scores return as reusable system context.

Reference architecture

Six layers. One familiar API outside.

01InterfaceOpenAI · Anthropic · Ollama · existing clients
02IntelligenceMeta-orchestrator · planner · templates · routing · policies
03ExecutionLLMs · SLMs · expert workers · agentic_tool · MCP · research
04KnowledgeNeo4j GraphRAG · vectors · documents · cache · memory
05QualityJudge · Self-Correction · Corrective RAG · security gates
06ObservabilityTracing · metrics · audit · routing telemetry

EuroHPC · LUMI-G

Use supercomputers so the platform needs fewer supercomputers later.

One-time intensive compute supports research into planning, routing and validation. The deployment target is smaller local specialist models—not permanent HPC dependence.

Support is provided by the EuroHPC Joint Undertaking as an allocation of high-performance computing resources, not as a direct financial grant.

View research documentation →
ImplementedCompound-AI pipelineTemplates, GraphRAG, tools, validation, memory
In validationTraining and evaluation dataTraceable plans and quality decisions
In developmentLUMI-G distillationComplexity, routing, tool selection, validation
PlannedSpecialised SLMsMore local handling of recurring tasks

Research corner

Test hypotheses—not just demonstrate systems.

The research asks which parts of a compound-AI system can measurably learn, when persistent knowledge can offset model size, and how quality can be assessed transparently.

Q₀

Can a learning orchestration and knowledge architecture enable small, locally deployable models to handle demanding tasks with improving quality and more efficient repeat processing?

R1
In development

Routing & plan distillation

Hypothesis: Successful expert plans can be transferred into smaller routers and planners without asking a large model to re-plan every request.

Method
Supervised fine-tuning, routing evaluation and ablations on LUMI-G
Measures
Plan quality, routing accuracy, latency and resource use
R2
In validation

Trust metrics & Judge

Hypothesis: Source, graph and contradiction signals can support more reliable quality decisions than model self-assessment alone.

Method
Paraconsistent conflict assessment, trust scores and controlled comparisons
Measures
Factuality, conflict detection, calibration and error rates
R3
System basis active

Knowledge accumulation

Hypothesis: Verified facts, solution paths and corrections shorten similar future processing and reduce repeated model work.

Method
GraphRAG, episodic memory, semantic caches and repeated task series
Measures
Cache/knowledge hits, latency, tokens, reuse and answer quality

Methodological guardrails

Reproducible, status-clear, falsifiable.

  • Document configuration, version and hardware
  • Separate internal measures from external baselines
  • Report sample, variance and limitations
  • Do not hide negative or failed results
01Scientific preprint12 pages · Preview PDF02Complete whitepaper108 pages · Preview PDF03Benchmark artefactsCode and methodology ↗04Research documentationArchitecture and setup ↗

From prompt to retained knowledge

Concrete workflows, not industry labels.

Public administrationReview procurement documents against current rules
“Check this specification for contradictions and missing evidence.”
Template
Document analysis → legal sources → contradiction check
Resources
Local LLM, documents, GraphRAG, current official sources
Validation
Source rank, citations, Judge, uncertainty labels
Retained
Approved terms, relations, corrections and review patterns
IT & securityAssess an architecture against threats
“Build a threat model for this hybrid platform and prioritise controls.”
Template
System model → threat research → controls → review
Resources
Security expert, MCP, internal standards, current advisories
Validation
Coverage, source recency, contradictions, Judge
Retained
System relations, accepted risks and validated controls
Software engineeringSolve a reproducible production fault
“Find the cause of this intermittent failure and provide a tested patch.”
Template
Reproduce → analyse code → patch → test → review
Resources
Code SLM/LLM, repository tools, tests, documentation
Validation
Build, tests, static analysis, diff review
Retained
Root cause, verified solution path and relevant code relations

Evidence, not superlatives

Metrics with context.

Internal benchmarks describe one concrete configuration. They are not universal claims against frontier models or every task.

9.3×

Accumulation latency

707 s → 76 s across five internal epochs through reuse of knowledge and cache paths.

Internal benchmark · April 2026 · small sample; hardware- and task-specificMethod and artefacts ↗
14/30

GAIA Level 1

46.7% in one documented internal run over 30 tasks.

Internal run · April 2026 · not representative of the full benchmark; consider setup and varianceInspect benchmark ↗

Sovereignty & security

Control is a technical property—not a blanket legal claim.

MoE Sovereign provides technical prerequisites for controllable data flows and data protection by design. Legal assessment depends on the actual deployment, purpose and organisational controls.

Review the versioned privacy self-assessment →
  • Local inference and optional offline/air-gap operation
  • Explicitly configurable outbound connections
  • Versioned Expert Templates and configuration
  • Routing, request and audit telemetry
  • Tenant, role and budget controls
  • OCI-portable and rootless operation
  • No required hidden product telemetry
  • Security gates and hardened tool paths

Deploy to fit

From local system to platform.

An OCI-based architecture supports different operating models. Kubernetes and OpenShift are architecturally prepared; production validation remains deployment-specific.

ONE-LINE INSTALLER · LINUX

Install MoE Sovereign interactively

For a fresh Debian or Ubuntu server. The installer detects Docker or Podman, generates secrets and guides the configuration.

curl -sSL https://moe-sovereign.org/install.sh | bash

Review shell scripts before execution: inspect install.sh · requirements and manual installation

Solo

One host

Local models, private knowledge base, full control.

Team

Shared instance

Central API, roles, templates and knowledge spaces.

Enterprise

Scaled platform

Tenancy, observability and integrated policies.

LXC · Docker Compose · Podman Quadlet · Kubernetes · OpenShift

The architecture is open

Understand more than the model.
Understand the process behind it.

Read the whitepaperView repositoryDeploy locally →