Update app.py

app.py

@@ -1,23 +1,15 @@
 import gradio as gr
-from llama_cpp import Llama
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
-# Download het biomedische GGML-model (direct van Hugging Face)
-model_url = "https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGML"
+model_id = "raidium/MQG"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id)
 
-# Initialiseer het model (werkt op CPU)
-llm = Llama(
-    model_path=model_url,
-    n_ctx=2048,
-    n_threads=4,
-)
+def answer(message, history=[]):
+    inputs = tokenizer.encode(message, return_tensors="pt")
+    outputs = model.generate(inputs, max_length=256)
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return response
 
-def chat_fn(message, history):
-    output = llm(
-        f"Je bent een biomedische chatbot. Beantwoord de vraag: {message}",
-        max_tokens=200,
-        temperature=0.7,
-    )
-    return output['choices'][0]['text']
-
-iface = gr.ChatInterface(chat_fn, title="Biomedische Chatbot")
+iface = gr.ChatInterface(answer, title="Medische Q&A (MQG)")
 iface.launch()