@@ -12,18 +12,23 @@ a ready‑made image in your own infrastructure.
1212 and optional Prometheus metrics.
1313- ** llm_router_lib** is a Python SDK that wraps the API with typed request/response models, automatic retries, token
1414 handling and a rich exception hierarchy, letting developers focus on application logic rather than raw HTTP calls.
15- - ** llm_router_web** offers ready‑to‑use Flask UIs – an anonymizer UI that masks sensitive data and a configuration
15+ - [ ** llm_router_web** ] ( https://github.com/radlab-dev-group/llm-router-web ) offers ready‑to‑use Flask UIs – an anonymizer
16+ UI that masks sensitive data and a configuration
1617 manager for model/user settings – demonstrating how to consume the router from a browser.
17- - ** llm_router_plugins** (e.g., the ** fast_masker** plugin) deliver a rule‑based text anonymisation engine with
18+ - [ ** llm_router_plugins** ] ( https://github.com/radlab-dev-group/llm-router-plugins ) (e.g., the ** fast_masker** plugin)
19+ deliver a rule‑based text anonymisation engine with
1820 a comprehensive set of Polish‑specific masking rules (emails, IPs, URLs, phone numbers, PESEL, NIP, KRS, REGON,
1921 monetary amounts, dates, etc.) and an extensible architecture for custom rules and validators.
22+ - [ ** llm_router_services** ] ( https://github.com/radlab-dev-group/llm-router-services ) provides HTTP services that
23+ implement the core functionality used by the LLM‑Router’s plugin system. The services expose guardrail and masking
24+ capabilities through Flask applications.
2025
2126All components run on Python 3.10+ using ` virtualenv ` and require only the listed dependencies, making the suite easy to
2227install, extend, and deploy in both development and production environments.
2328
2429---
2530
26- ### ✨ Key Features
31+ ## ✨ Key Features
2732
2833| Feature | Description |
2934| -------------------------------------| ---------------------------------------------------------------------------------------------------------------------------------------------------------|
@@ -52,21 +57,6 @@ install, extend, and deploy in both development and production environments.
5257
5358#### Base requirements
5459
55- > ** Prerequisite** : ` radlab-ml-utils `
56- >
57- > This project uses the
58- > [ radlab-ml-utils] ( https://github.com/radlab-dev-group/ml-utils )
59- > library for machine learning utilities
60- > (e.g., experiment/result logging with Weights & Biases/wandb).
61- > Install it before working with ML-related parts:
62- >
63- > ``` bash
64- > pip install git+https://github.com/radlab-dev-group/ml-utils.git
65- > ` ` `
66- >
67- > For more options and details, see the library README:
68- > https://github.com/radlab-dev-group/ml-utils
69-
7060``` shell script
7161python3 -m venv .venv
7262source .venv/bin/activate
@@ -108,7 +98,7 @@ metrics for monitoring and alerting.
10898LLM_ROUTER_MINIMUM=1 python3 -m llm_router_api.rest_api
10999```
110100
111- ### 📦 Docker
101+ ## 📦 Docker
112102
113103Run the container with the default configuration:
114104
@@ -147,7 +137,7 @@ docker run \
147137
148138---
149139
150- # ## Configuration (via environment)
140+ # # 🛠️ Configuration (via environment)
151141
152142A full list of environment variables is available at the link
153143[.env list](llm_router_api/README.md#environment-variables)
@@ -184,7 +174,7 @@ a description of the streaming mechanisms can be found at the link:
184174
185175---
186176
187- # # 🛠️ Development
177+ # # 🔧 Development
188178
189179- ** Python** 3.10+ (project is tested on 3.10.6)
190180- All dependencies are listed in ` requirements.txt` . Install them inside the virtualenv.
0 commit comments