Create app.py

app.py

@@ -0,0 +1,33 @@
+import os
+import gradio as gr
+from transformers import AutoTokenizer, AutoModelForCausalLM, TextStreamer
+
+model_id = os.getenv("MODEL_ID", "mistralai/Mistral-7B-Instruct-v0.3")
+
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(
+    model_id,
+    device_map="auto",
+    load_in_4bit=True,
+    trust_remote_code=True
+)
+
+streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
+
+def generate_response(message, history):
+    messages = history + [[message, ""]]
+    prompt = ""
+    for user, bot in messages:
+        prompt += f"<|user|>\n{user}\n<|assistant|>\n{bot}\n"
+    prompt += f"<|user|>\n{message}\n<|assistant|>\n"
+
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    output = model.generate(**inputs, max_new_tokens=512, streamer=streamer)
+    response = tokenizer.decode(output[0], skip_special_tokens=True).split("<|assistant|>")[-1].strip()
+
+    return response
+
+demo = gr.ChatInterface(fn=generate_response, title="Vynix AI", theme="soft")
+
+if _name_ == "_main_":
+    demo.launch()