Create app.py

app.py

@@ -0,0 +1,18 @@
+import gradio as gr
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
+
+model_id = "BioMistral/BioMistral-7B"  # Of een clinical variant als je dat wilt
+
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id)
+pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
+
+def biomistral_chat(message, history):
+    prompt = f"### Question:\n{message}\n### Answer:\n"
+    response = pipe(prompt, max_new_tokens=256)[0]["generated_text"]
+    # Haal alleen het antwoorddeel eruit
+    answer = response.split("### Answer:")[-1].strip()
+    return answer
+
+demo = gr.ChatInterface(biomistral_chat)
+demo.launch()