Small improvement.

Author: Vadim Nicolai

README.md

@@ -39,4 +39,87 @@ To learn more about the AI SDK, Next.js, and FastAPI take a look at the followin
 - [Vercel AI Playground](https://play.vercel.ai) - try different models and choose the best one for your use case.
 - [Next.js Docs](https://nextjs.org/docs) - learn about Next.js features and API.
 - [FastAPI Docs](https://fastapi.tiangolo.com) - learn about FastAPI features and API.
+
 # email-finder
+
+Below is a brief README-style explanation for running DeepSeek-R1 locally with Ollama.
+
+---
+
+# DeepSeek-R1: Local Setup Guide
+
+This guide explains how to run DeepSeek-R1 on your local machine using [Ollama](https://ollama.ai).
+
+## 1. Install Ollama
+
+1. **Download**: Visit the [Ollama website](https://ollama.ai) and download the installer for your operating system.
+2. **Install**: Install Ollama as you would any other application.
+
+## 2. Download and Test DeepSeek-R1
+
+1. **Open Terminal**: Launch your terminal or command prompt.
+2. **Run the Model**:
+
+   ```bash
+   ollama run deepseek-r1
+   ```
+
+   This command automatically downloads the DeepSeek-R1 model (default size) and runs a sample prompt.
+
+3. **Alternate Model Sizes** (optional):
+   ```bash
+   ollama run deepseek-r1:<size>b
+   ```
+   Replace `<size>` with `1.5`, `7`, `8`, `14`, `32`, `70`, or `671` to download/run smaller or larger versions.
+
+## 3. Run DeepSeek-R1 as a Service
+
+To keep DeepSeek-R1 running in the background and serve requests via an API:
+
+```bash
+ollama serve
+```
+
+This exposes DeepSeek-R1 at `http://localhost:11434/api/chat` for integration with other applications.
+
+## 4. Test via CLI and API
+
+- **CLI**: Once DeepSeek-R1 is running, simply type:
+  ```bash
+  ollama run deepseek-r1
+  ```
+- **API**: Use `curl` to chat with DeepSeek-R1 via the local server:
+  ```bash
+  curl http://localhost:11434/api/chat -d '{
+    "model": "deepseek-r1",
+    "messages": [{ "role": "user", "content": "Hello DeepSeek, how are you?" }],
+    "stream": false
+  }'
+  ```
+
+## 5. Next Steps
+
+- **Python Integration**: Use the `ollama` Python package to integrate DeepSeek-R1 into applications:
+
+  ```python
+  import ollama
+
+  response = ollama.chat(
+      model="deepseek-r1",
+      messages=[{"role": "user", "content": "Hi DeepSeek!"}],
+  )
+  print(response["message"]["content"])
+  ```
+
+- **Gradio App**: Build a simple web interface (e.g., for RAG tasks) using [Gradio](https://gradio.app).
+
+For more details on prompt construction, chunk splitting, or building retrieval-based applications (RAG), refer to the official documentation and tutorials.
+
+---
+
+## References
+
+- [Ollama Documentation](https://ollama.ai)
+- [DeepSeek-R1 Article](#) (replace `#` with your desired URL if available)
+
+---