diff --git a/transcription/.gitignore b/transcription/.gitignore
new file mode 100644
index 00000000..e69de29b
diff --git a/transcription/app.py b/transcription/app.py
new file mode 100644
index 00000000..bff6161e
--- /dev/null
+++ b/transcription/app.py
@@ -0,0 +1,40 @@
+from flask import Flask, request, jsonify
+from PIL import Image
+from transformers import AutoProcessor, AutoModelForCausalLM
+import torch
+
+app = Flask(__name__)
+
+# load model and processor once during init
+device = "cuda:0" if torch.cuda.is_available() else "cpu"
+torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+
+model = AutoModelForCausalLM.from_pretrained("microsoft/Florence-2-large", torch_dtype=torch_dtype, trust_remote_code=True).to(device)
+processor = AutoProcessor.from_pretrained("microsoft/Florence-2-large", trust_remote_code=True)
+
+@app.route("/transcribe", methods=["POST"])
+def transcribe():
+    if "image" not in request.files:
+        return jsonify({"error": "No image file provided"}), 400
+
+    image_file = request.files["image"]
+    try:
+        # open and preprocess image
+        image = Image.open(image_file).convert("RGB")
+        prompt = "<OCR>"
+        inputs = processor(text=prompt, images=image, return_tensors="pt").to(device, torch_dtype)
+        generated_ids = model.generate(
+            input_ids=inputs["input_ids"],
+            pixel_values=inputs["pixel_values"],
+            max_new_tokens=1024,
+            num_beams=3,
+            do_sample=False
+        )
+        generated_text = processor.batch_decode(generated_ids, skip_special_tokens=False)[0]
+
+        return jsonify({"transcription": generated_text})
+    except Exception as e:
+        return jsonify({"error": str(e)}), 500
+
+if __name__ == "__main__":
+    app.run(debug=True)
\ No newline at end of file
diff --git a/transcription/requirements.txt b/transcription/requirements.txt
index 03feafad..d87174ed 100644
--- a/transcription/requirements.txt
+++ b/transcription/requirements.txt
@@ -5,3 +5,5 @@ torch==2.5.1
 transformers==4.46.1
 einops
 timm
+flask
+flask-cors
\ No newline at end of file
diff --git a/transcription/testing.py b/transcription/testing.py
index f569788a..499e0d66 100644
--- a/transcription/testing.py
+++ b/transcription/testing.py
@@ -33,7 +33,7 @@ def florence():
 
     prompt = "<OCR>"
 
-    url = "../assets/kkl.jpg"
+    url = "../assets/Filled_Logbook_page-0001.jpg"
     image = Image.open(url).convert("RGB")
 
     inputs = processor(text=prompt, images=image, return_tensors="pt").to(device, torch_dtype)