> ## Documentation Index
> Fetch the complete documentation index at: https://doc.blueapi.ir/llms.txt
> Use this file to discover all available pages before exploring further.

# EMEP: Business and Technical Context

> Understand who EMEP serves, what triggered the project, and where it sits among open-weight LLM tooling like MergeKit and LM Evaluation Harness.

EMEP is a documentation-first platform for building new open-weight large language models through compatibility analysis, tensor-level merging, evaluation, and evolutionary optimization. This page defines the audience, origin, and competitive landscape.

## Audience

EMEP targets three primary user groups:

* **Research engineers** experimenting with model merging and evolutionary optimization
* **MLOps teams** operating model registries, benchmarks, and deployment pipelines
* **Enterprise practitioners** requiring offline-capable, auditable model engineering

## Origin

The project was triggered by the rapid proliferation of open-weight models and the absence of an integrated platform that combines merge operations, compatibility validation, evolutionary search, and reproducible evaluation. Existing tools solve fragments of this pipeline. EMEP unifies them under a single specification.

## Competitive Landscape

```mermaid theme={null}
flowchart TB
    subgraph Open["Open-Weight Ecosystem"]
        A["Base Models<br/>Llama, Mistral, Qwen"]
        B["MergeKit<br/>Task Arithmetic, TIES, DARE"]
        C["LM Evaluation Harness<br/>Benchmark execution"]
        D["Hugging Face Hub<br/>Model distribution"]
    end

    subgraph EMEP["EMEP"]
        E["Compatibility Analysis"]
        F["Tensor Merge Engine"]
        G["Evolutionary Optimization"]
        H["Evaluation Framework"]
        I["Registry + Tracking"]
    end

    subgraph Closed["Closed APIs"]
        J["OpenAI API"]
        K["Anthropic API"]
    end

    A --> B
    A --> C
    A --> D
    B --> EMEP
    C --> EMEP
    D --> EMEP
    EMEP --> J
    EMEP --> K

    style EMEP fill:#e1f5fe
    style E fill:#e1f5fe
    style F fill:#e1f5fe
    style G fill:#e1f5fe
    style H fill:#e1f5fe
    style I fill:#e1f5fe
```

EMEP sits between raw model repositories and closed APIs. It consumes base models from the open ecosystem, applies structured merge and evolution operations, and produces candidate models ready for evaluation, quantization, and deployment.

## What EMEP Does Not Replace

* **MergeKit**: EMEP implements its own [MergeEngine](/merge/merge-engine) with compatible strategies, but does not depend on MergeKit as a library
* **LM Evaluation Harness**: EMEP defines its own [EvaluationEngine](/evaluation/framework) with benchmark specifications that may call Harness tasks, but maintains independent result tracking
* **OpenAI / Anthropic APIs**: EMEP produces open-weight models. It does not provide hosted inference APIs
* **Hugging Face Hub**: EMEP may import from and export to the Hub, but maintains its own [ModelRegistry](/tracking/model-registry) and [ArtifactStore](/tracking/artifact-registry)

## Key Differentiators

* **Documentation-first**: Every feature is specified before implementation
* **Evolutionary optimization**: Multi-objective search over merge strategies, not single-shot composition
* **Offline-first enterprise mode**: Full operation without external network dependencies
* **Hidden Test Set isolation**: [Benchmark splits](/evaluation/benchmark-specification) enforce strict separation between optimization and final evaluation
* **Reproducibility protocol**: Every experiment is tracked, signed, and lineage-linked

## Next Steps

Read the [Vision and Scope](/overview/vision-and-scope) for boundaries and phase overview, or the [Project Constitution](/overview/project-constitution) for non-negotiable scientific rules.