# ilirium/ilirium_llm_router architecture

> Architecture diagram of the GitHub repository ilirium/ilirium_llm_router, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-01). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/ilirium/ilirium_llm_router
- Repository: https://github.com/ilirium/ilirium_llm_router

## Overview

ilirium_llm_router is an Anthropic-compatible proxy that lets Claude Code use Anthropic and LM Studio in one session. The main workflow is a client request entering the FastAPI app, being forwarded by the proxy to the model-selected backend, and returning as a streamed response. Calls also feed telemetry and, when enabled, an opaque body corpus. CLI commands configure and serve the router, train compression dictionaries, and extract or verify archived calls. Edges reflect README statements and shown source relationships; unsampled implementation details are not inferred.

Main entry points: [`src/ilirium_llm_router/proxy.py`](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/proxy.py) (Request proxy), [`src/ilirium_llm_router/cli.py`](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/cli.py) (CLI commands), [`src/ilirium_llm_router/app.py`](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/app.py) (HTTP application), [`src/ilirium_llm_router/routing.py`](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/routing.py) (Model routing)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_entry["Entry and configuration"]
  node_cli["CLI commands<br/>[cli.py]"]
  node_config["Configuration<br/>[config.py]"]
  node_app["HTTP application<br/>[app.py]"]
end

subgraph group_proxy["Request routing"]
  node_proxy["Request proxy<br/>[proxy.py]"]
  node_routing["Model routing<br/>[routing.py]"]
end

subgraph group_observability["Call observability"]
  node_observe["Usage measurement<br/>[observe.py]"]
  node_stats[("Telemetry CSV<br/>[stats.py]")]
  node_logging[("Rotating logs<br/>[logging_setup.py]")]
end

subgraph group_archive["Corpus and replay"]
  node_corpus[("Body corpus<br/>[corpus.py]")]
  node_dictionary["Dictionary training<br/>[dictionary.py]"]
  node_extract["Archive extraction<br/>[extract.py]"]
  node_jsonl["JSONL output<br/>[jsonl.py]"]
  node_transcript["Transcript replay<br/>[transcript.py]"]
end

node_claude(("Claude Code"))
node_anthropic{{"Anthropic API"}}
node_lmstudio{{"LM Studio"}}

node_claude -->|"sends requests"| node_app
node_app -->|"dispatches routes"| node_proxy
node_proxy -->|"selects backend"| node_routing
node_routing -->|"selects claude models"| node_anthropic
node_routing -->|"selects other models"| node_lmstudio
node_proxy -->|"forwards requests"| node_anthropic
node_proxy -->|"forwards requests"| node_lmstudio
node_anthropic -->|"streams replies"| node_proxy
node_lmstudio -->|"streams replies"| node_proxy
node_proxy -->|"measures calls"| node_observe
node_observe -->|"records metrics"| node_stats
node_proxy -->|"writes call rows"| node_stats
node_app -->|"sets up logging"| node_logging
node_proxy -->|"captures bodies"| node_corpus
node_corpus -->|"supplies training data"| node_dictionary
node_cli -->|"loads settings"| node_config
node_cli -->|"starts server"| node_app
node_cli -->|"runs training"| node_dictionary
node_cli -->|"selects archived calls"| node_extract
node_extract -->|"writes JSONL"| node_jsonl
node_extract -->|"replays replies"| node_transcript
node_jsonl -->|"uses replay logic"| node_transcript
node_cli -->|"reads archive"| node_corpus

click node_cli "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/cli.py"
click node_config "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/config.py"
click node_app "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/app.py"
click node_proxy "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/proxy.py"
click node_routing "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/routing.py"
click node_observe "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/observe.py"
click node_stats "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/stats.py"
click node_logging "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/logging_setup.py"
click node_corpus "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/corpus.py"
click node_dictionary "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/dictionary.py"
click node_extract "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/extract.py"
click node_jsonl "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/jsonl.py"
click node_transcript "https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/transcript.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_cli,node_config,node_app toneBlue
class node_proxy,node_routing toneAmber
class node_observe,node_stats,node_logging,node_anthropic toneMint
class node_corpus,node_dictionary,node_extract,node_jsonl,node_transcript toneRose
class node_claude,node_lmstudio toneIndigo
```

## Components

### Entry and configuration

- [CLI commands](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/cli.py): `src/ilirium_llm_router/cli.py`
- [Configuration](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/config.py): `src/ilirium_llm_router/config.py`
- [HTTP application](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/app.py): `src/ilirium_llm_router/app.py`

### Request routing

- [Request proxy](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/proxy.py): `src/ilirium_llm_router/proxy.py`
- [Model routing](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/routing.py): `src/ilirium_llm_router/routing.py`

### Call observability

- [Usage measurement](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/observe.py): `src/ilirium_llm_router/observe.py`
- [Telemetry CSV](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/stats.py): `src/ilirium_llm_router/stats.py`
- [Rotating logs](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/logging_setup.py): `src/ilirium_llm_router/logging_setup.py`

### Corpus and replay

- [Body corpus](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/corpus.py): `src/ilirium_llm_router/corpus.py`
- [Dictionary training](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/dictionary.py): `src/ilirium_llm_router/dictionary.py`
- [Archive extraction](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/extract.py): `src/ilirium_llm_router/extract.py`
- [JSONL output](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/jsonl.py): `src/ilirium_llm_router/jsonl.py`
- [Transcript replay](https://github.com/ilirium/ilirium_llm_router/blob/main/src/ilirium_llm_router/transcript.py): `src/ilirium_llm_router/transcript.py`

### Other components

- **Claude Code**
- **Anthropic API**
- **LM Studio**

## Connections

- Claude Code → HTTP application: sends requests (evidence: `README.md`)
- HTTP application → Request proxy: dispatches routes (evidence: `src/ilirium_llm_router/app.py`)
- Request proxy → Model routing: selects backend (evidence: `src/ilirium_llm_router/proxy.py`)
- Model routing → Anthropic API: selects claude models (evidence: `README.md`)
- Model routing → LM Studio: selects other models (evidence: `README.md`)
- Request proxy → Anthropic API: forwards requests (evidence: `README.md`)
- Request proxy → LM Studio: forwards requests (evidence: `README.md`)
- Anthropic API → Request proxy: streams replies (evidence: `README.md`)
- LM Studio → Request proxy: streams replies (evidence: `README.md`)
- Request proxy → Usage measurement: measures calls (evidence: `src/ilirium_llm_router/proxy.py`)
- Usage measurement → Telemetry CSV: records metrics (evidence: `src/ilirium_llm_router/observe.py`)
- Request proxy → Telemetry CSV: writes call rows (evidence: `src/ilirium_llm_router/app.py`)
- HTTP application → Rotating logs: sets up logging (evidence: `src/ilirium_llm_router/cli.py`)
- Request proxy → Body corpus: captures bodies (evidence: `src/ilirium_llm_router/proxy.py`)
- Body corpus → Dictionary training: supplies training data (evidence: `src/ilirium_llm_router/dictionary.py`)
- CLI commands → Configuration: loads settings (evidence: `src/ilirium_llm_router/cli.py`)
- CLI commands → HTTP application: starts server (evidence: `src/ilirium_llm_router/cli.py`)
- CLI commands → Dictionary training: runs training (evidence: `src/ilirium_llm_router/cli.py`)
- CLI commands → Archive extraction: selects archived calls (evidence: `src/ilirium_llm_router/cli.py`)
- Archive extraction → JSONL output: writes JSONL (evidence: `src/ilirium_llm_router/extract.py`)
- Archive extraction → Transcript replay: replays replies (evidence: `src/ilirium_llm_router/extract.py`)
- JSONL output → Transcript replay: uses replay logic (evidence: `src/ilirium_llm_router/jsonl.py`)
- CLI commands → Body corpus: reads archive (evidence: `src/ilirium_llm_router/cli.py`)
