add fast api app

app.py

@@ -0,0 +1,33 @@
+from fastapi import FastAPI
+from pydantic import BaseModel
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
+
+app = FastAPI()
+
+# Load the Mongolian Llama model and tokenizer
+model_name = "Dorjzodovsuren/Mongolian_Llama3-v0.1"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name)
+
+class UserInput(BaseModel):
+    text: str
+
+@app.post("/generate/")
+def generate_response(user_input: UserInput):
+    # Tokenize the input text
+    inputs = tokenizer(user_input.text, return_tensors="pt")
+    
+    # Generate response
+    with torch.no_grad():
+        outputs = model.generate(
+            **inputs,
+            max_length=100,  # Adjust for desired response length
+            num_return_sequences=1,
+            temperature=0.7,  # Adjust for creativity
+            top_p=0.9        # Adjust for response diversity
+        )
+    
+    # Decode the generated text
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return {"response": response}

requirements.txt

@@ -0,0 +1,4 @@
+fastapi
+uvicorn
+transformers
+torch