diff --git a/README.md b/README.md
index 703fcfc..54fae1c 100644
--- a/README.md
+++ b/README.md
@@ -1,3 +1,72 @@
 # VeraMind
 
-Open Weights Fake News Detection Model and Inference
\ No newline at end of file
+
+The VeraMind is an open-source Python application built using the Hugging Face Transformers library and PyTorch. It leverages a pre-trained model (`VeraMind-Mini`) to predict whether a given news article is real or fake with a confidence score.
+
+This project is licensed under the [Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)](https://creativecommons.org/licenses/by-nc-nd/4.0/) license. You are free to use and share this model privately, but you must give appropriate credit, not use it for commercial purposes, and not distribute derivative works.
+
+**Note:** This is a machine learning model and may make mistakes. It should not replace your own critical thinking when evaluating news authenticity. Always verify information from multiple reliable sources.
+
+## Features
+
+- Predicts if a given news article is real or fake.
+- Provides a confidence score for the prediction.
+- Utilizes the Hugging Face Transformers library for easy integration with other NLP models.
+
+## Installation
+
+1. Clone this repository:
+
+```bash
+git clone https://github.com/yourusername/VeraMind.git
+cd VeraMind
+```
+
+2. Install the required dependencies:
+
+```bash
+pip install -r requirements.txt
+```
+
+## Usage
+
+### Predicting News Authenticity
+
+Here's how you can use the model to predict if a news article is real or fake:
+
+```python
+from src.Inference import VeraMindInference
+
+# Load the model
+model = VeraMindInference("path/to/VeraMind-Mini")
+
+# Example news article text
+text = "This is an example News Article"
+
+# Predict if the news is real or fake
+result = model.predict(text)
+
+print(result)
+```
+
+The output will be a dictionary containing the result ("REAL" or "FAKE") and the confidence score:
+
+```python
+{'result': 'FAKE', 'confidence': 0.9990140199661255}
+```
+
+## Model Architecture
+
+The `VeraMind-Mini` model used in this application is a fine-tuned version of the [DistilBERT](https://huggingface.co/distilbert-base-uncased) model for binary text classification. It's designed to distinguish between real and fake news articles.
+
+## Disclaimer
+
+This project is provided as-is, without any express or implied warranty. The maintainers are not responsible for any damages arising from the use of this software.
+
+Always remember that machine learning models can make mistakes, so use this tool responsibly and critically evaluate its predictions.
+
+## Citation
+
+If you use this model in your research, please cite it as follows:
+
+> **VeraMind News Authenticity Checker** (2024). Retrieved from https://gitea.fabelous.app/Fabel/VeraMind by Falko Habel
\ No newline at end of file
diff --git a/main.py b/main.py
new file mode 100644
index 0000000..62dcd04
--- /dev/null
+++ b/main.py
@@ -0,0 +1,15 @@
+from src.Inference import VeraMindInference
+
+
+# load model
+model = VeraMindInference("path/to/VeraMind-Mini")
+
+
+text = "This is a example News Article"
+
+# predict if News are reel or Fake
+result = model.predict(text)
+
+# Example Output
+# {'result': 'FAKE', 'confidence': 0.9990140199661255}
+print(result)
diff --git a/requirements.txt b/requirements.txt
new file mode 100644
index 0000000..9216a18
--- /dev/null
+++ b/requirements.txt
@@ -0,0 +1,2 @@
+torch
+transformers
\ No newline at end of file
diff --git a/src/Inference.py b/src/Inference.py
new file mode 100644
index 0000000..6814509
--- /dev/null
+++ b/src/Inference.py
@@ -0,0 +1,38 @@
+import torch
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+
+class VeraMindInference:
+    def __init__(self, model_path, max_len=512):
+        self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+        self.tokenizer = AutoTokenizer.from_pretrained(model_path)
+        self.model = AutoModelForSequenceClassification.from_pretrained(model_path)
+        self.model.to(self.device)
+        self.model.eval()
+        self.max_len = max_len
+
+    def predict(self, text):
+        encoding = self.tokenizer.encode_plus(
+            text,
+            add_special_tokens=True,
+            max_length=self.max_len,
+            return_token_type_ids=False,
+            padding='max_length',
+            truncation=True,
+            return_attention_mask=True,
+            return_tensors='pt',
+        )
+
+        input_ids = encoding['input_ids'].to(self.device)
+        attention_mask = encoding['attention_mask'].to(self.device)
+
+        with torch.no_grad():
+            outputs = self.model(input_ids, attention_mask=attention_mask).logits
+            prediction = torch.sigmoid(outputs).cpu().numpy()[0][0]
+
+        is_fake = prediction >= 0.5
+        confidence = prediction if is_fake else 1 - prediction
+
+        return {
+            "result": "FAKE" if is_fake else "REAL",
+            "confidence": float(confidence)
+        }
\ No newline at end of file
diff --git a/src/__init__.py b/src/__init__.py
new file mode 100644
index 0000000..e69de29