From f7dde68e9eae3142f2e3d4fc2c9aee185ffe060b Mon Sep 17 00:00:00 2001
From: michaelmanley <55236695+michaelbrinkworth@users.noreply.github.com>
Date: Wed, 10 Dec 2025 09:28:40 +1000
Subject: [PATCH] docs: add AI Badgr as OpenAI-compatible backend (base_url
 override + examples)

---
 README.md | 33 +++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/README.md b/README.md
index 5dceac7..ecceb69 100644
--- a/README.md
+++ b/README.md
@@ -15,3 +15,36 @@ $ cd Rebel-OpenAI
 $ npm install
 $ node index.js</div>
 ```
+
+### Using AI Badgr (OpenAI-Compatible Backend)
+
+If you're already using the OpenAI API, you can also point your client at **AI Badgr**, which implements the same API surface (chat completions, streaming, JSON mode).
+
+AI Badgr uses optimized open-source models to provide more affordable inference and also supports optional private/self-hosted modes for teams that need additional data control.
+
+```bash
+export OPENAI_API_KEY=YOUR_API_KEY
+export OPENAI_BASE_URL=https://aibadgr.com/api/v1
+```
+
+**JavaScript/TypeScript:**
+```javascript
+import OpenAI from 'openai';
+const client = new OpenAI({ apiKey: 'YOUR_API_KEY', baseURL: 'https://aibadgr.com/api/v1' });
+const response = await client.chat.completions.create({ model: 'gpt-3.5-turbo', messages: [{ role: 'user', content: 'Hello!' }], max_tokens: 200 });
+console.log(response.choices[0].message.content);
+```
+
+
+**cURL:**
+```bash
+curl https://aibadgr.com/api/v1/chat/completions \
+  -H "Authorization: Bearer YOUR_API_KEY" \
+  -H "Content-Type: application/json" \
+  -d '{"model":"gpt-3.5-turbo","messages":[{"role":"user","content":"Hello!"}],"max_tokens":200}'
+```
+
+**Notes:**
+- Streaming: `"stream": true`
+- JSON mode: `"response_format": {"type": "json_object"}`
+- Because AI Badgr follows the OpenAI API spec, most existing OpenAI client code works by only changing the `base_url`.