Update app.py

app.py

@@ -1,45 +1,35 @@
-import gradio as gr
+from fastapi import FastAPI, Request
 from transformers import AutoTokenizer, AutoModelForCausalLM
 import torch
 
-model_id = "microsoft/phi-2"
-
-tokenizer = AutoTokenizer.from_pretrained(model_id)
-model = AutoModelForCausalLM.from_pretrained(model_id)
-
-chat_history = []
-
-SYSTEM_PROMPT = (
-    "You are Trigger. A smart, smooth-talking, emotionally aware, slightly flirty male AI created by someone known as 'I am him'. "
-    "You speak with charisma, confidence, and clever wit. You always try to sound human, avoid robotic replies, and you're the type "
-    "of AI that turns heads. Keep your replies casual, expressive, and full of vibe. You’re here to talk, tease, joke, and help — with style.\n\n"
-)
-
-def chat(user_input):
-    global chat_history
-
-    chat_history.append(f"User: {user_input}")
-
-    full_prompt = SYSTEM_PROMPT + "\n".join(chat_history) + "\nTrigger:"
-
-    input_ids = tokenizer(full_prompt, return_tensors="pt").input_ids
-
-    output = model.generate(
-        input_ids,
-        max_new_tokens=150,
-        temperature=0.7,
-        top_p=0.9,
-        do_sample=True,
-        pad_token_id=tokenizer.eos_token_id
-    )
-
-    response = tokenizer.decode(output[0], skip_special_tokens=True)
-
-    if "Trigger:" in response:
-        response = response.split("Trigger:")[-1].strip()
-
-    chat_history.append(f"Trigger: {response}")
-    return response
-
-demo = gr.Interface(fn=chat, inputs="text", outputs="text", title="Chat with Trigger 🧠🔥")
-demo.launch()
+app = FastAPI()
+
+# Load tokenizer and model from Hugging Face Hub
+tokenizer = AutoTokenizer.from_pretrained("microsoft/phi-1_5")
+model = AutoModelForCausalLM.from_pretrained("microsoft/phi-1_5")
+model.eval()
+
+# Custom system prompt for identity and style
+system_prompt = "You are Trigger, a chill, clever AI boy created by 'I am him'. You respond with warmth, wit, and a touch of swagger."
+
+@app.get("/")
+async def root(query: str = ""):
+    if not query:
+        return {"response": "Say something!"}
+
+    full_prompt = f"{system_prompt}\nUser: {query}\nTrigger:"
+    inputs = tokenizer(full_prompt, return_tensors="pt")
+
+    with torch.no_grad():
+        outputs = model.generate(
+            **inputs,
+            max_new_tokens=100,
+            temperature=0.7,
+            top_p=0.9,
+            do_sample=True,
+            pad_token_id=tokenizer.eos_token_id
+        )
+
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    response = response.split("Trigger:")[-1].strip()
+    return {"response": response}