diff --git a/ARCHITECTURE.md b/ARCHITECTURE.md
index d2282110992..6894c4660f7 100644
--- a/ARCHITECTURE.md
+++ b/ARCHITECTURE.md
@@ -35,47 +35,34 @@ sequenceDiagram
```mermaid
graph TD
subgraph "Incoming Request"
- Client["Client Request"]
+ Client["POST /v1/chat/completions"]
end
subgraph "proxy/proxy_server.py"
- Endpoint["API Endpoints
/v1/chat/completions, /v1/messages, etc."]
+ Endpoint["chat_completion()"]
end
subgraph "proxy/auth/"
- Auth["user_api_key_auth.py
API Key, JWT, OAuth2"]
+ Auth["user_api_key_auth()"]
end
- subgraph "proxy/hooks/"
- Hooks["max_budget_limiter
parallel_request_limiter
cache_control_check"]
- end
-
- subgraph "proxy/*_endpoints/"
- LLMEndpoints["anthropic_endpoints/
vertex_ai_endpoints/
google_endpoints/
pass_through_endpoints/"]
+ subgraph "proxy/"
+ PreCall["litellm_pre_call_utils.py"]
+ RouteRequest["route_llm_request.py"]
end
subgraph "litellm/"
- Router["router.py
Load Balancing"]
- Main["main.py
completion()"]
- end
-
- subgraph "proxy/management_endpoints/"
- Management["key_management.py
team_management.py
model_management.py"]
- end
-
- subgraph "proxy/db/"
- DB["prisma_client.py
PostgreSQL/SQLite"]
+ Router["router.py"]
+ Main["main.py"]
end
Client --> Endpoint
Endpoint --> Auth
- Auth --> Hooks
- Hooks --> LLMEndpoints
- LLMEndpoints --> Router
+ Auth --> PreCall
+ PreCall --> RouteRequest
+ RouteRequest --> Router
Router --> Main
- Endpoint --> Management
- Management --> DB
- Hooks --> DB
+ Main --> Client
```
**Key files:**