Spaces:
Running
Running
Update README.md
Browse files
README.md
CHANGED
|
@@ -40,39 +40,12 @@ Intelligently routes requests to specialized models based on semantic understand
|
|
| 40 |
|
| 41 |
Our testing shows significant improvements in model accuracy through specialized routing.
|
| 42 |
|
|
|
|
| 43 |
|
| 44 |
## π οΈ Architecture Overview
|
| 45 |
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
Client[Client Request] --> Envoy[Envoy Proxy]
|
| 49 |
-
Envoy --> Router[Semantic Router ExtProc]
|
| 50 |
-
|
| 51 |
-
subgraph "Classification Modules"
|
| 52 |
-
direction LR
|
| 53 |
-
PII[PII Detector]
|
| 54 |
-
Jailbreak[Jailbreak Guard]
|
| 55 |
-
Category[Category Classifier]
|
| 56 |
-
Cache[Semantic Cache]
|
| 57 |
-
end
|
| 58 |
-
|
| 59 |
-
Router --> PII
|
| 60 |
-
Router --> Jailbreak
|
| 61 |
-
Router --> Category
|
| 62 |
-
Router --> Cache
|
| 63 |
-
|
| 64 |
-
PII --> Decision{Security Check}
|
| 65 |
-
Jailbreak --> Decision
|
| 66 |
-
Decision -->|Block| Block[Block Request]
|
| 67 |
-
Decision -->|Pass| Category
|
| 68 |
-
Category --> Models[Route to Specialized Model]
|
| 69 |
-
Cache -->|Hit| FastResponse[Return Cached Response]
|
| 70 |
-
|
| 71 |
-
Models --> Math[Math Model]
|
| 72 |
-
Models --> Creative[Creative Model]
|
| 73 |
-
Models --> Code[Code Model]
|
| 74 |
-
Models --> General[General Model]
|
| 75 |
-
```
|
| 76 |
|
| 77 |
## π― Use Cases
|
| 78 |
|
|
@@ -88,6 +61,8 @@ The router provides comprehensive monitoring through:
|
|
| 88 |
- **Prometheus Metrics**: Detailed routing statistics and performance data
|
| 89 |
- **Request Tracing**: Full visibility into routing decisions and performance
|
| 90 |
|
|
|
|
|
|
|
| 91 |
## π Documentation
|
| 92 |
|
| 93 |
For comprehensive documentation including detailed setup instructions, architecture guides, and API references, visit:
|
|
@@ -99,5 +74,4 @@ The documentation includes:
|
|
| 99 |
- **[Quick Start](https://llm-semantic-router.readthedocs.io/en/latest/getting-started/quick-start/)** - Get running in 5 minutes
|
| 100 |
- **[System Architecture](https://llm-semantic-router.readthedocs.io/en/latest/architecture/system-architecture/)** - Technical deep dive
|
| 101 |
- **[Model Training](https://llm-semantic-router.readthedocs.io/en/latest/training/training-overview/)** - How classification models work
|
| 102 |
-
- **[API Reference](https://llm-semantic-router.readthedocs.io/en/latest/api/router/)** - Complete API documentation
|
| 103 |
-
|
|
|
|
| 40 |
|
| 41 |
Our testing shows significant improvements in model accuracy through specialized routing.
|
| 42 |
|
| 43 |
+

|
| 44 |
|
| 45 |
## π οΈ Architecture Overview
|
| 46 |
|
| 47 |
+
|
| 48 |
+

|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
|
| 50 |
## π― Use Cases
|
| 51 |
|
|
|
|
| 61 |
- **Prometheus Metrics**: Detailed routing statistics and performance data
|
| 62 |
- **Request Tracing**: Full visibility into routing decisions and performance
|
| 63 |
|
| 64 |
+

|
| 65 |
+
|
| 66 |
## π Documentation
|
| 67 |
|
| 68 |
For comprehensive documentation including detailed setup instructions, architecture guides, and API references, visit:
|
|
|
|
| 74 |
- **[Quick Start](https://llm-semantic-router.readthedocs.io/en/latest/getting-started/quick-start/)** - Get running in 5 minutes
|
| 75 |
- **[System Architecture](https://llm-semantic-router.readthedocs.io/en/latest/architecture/system-architecture/)** - Technical deep dive
|
| 76 |
- **[Model Training](https://llm-semantic-router.readthedocs.io/en/latest/training/training-overview/)** - How classification models work
|
| 77 |
+
- **[API Reference](https://llm-semantic-router.readthedocs.io/en/latest/api/router/)** - Complete API documentation
|
|
|