Create main.py

main.py

@@ -0,0 +1,29 @@
+from fastapi import FastAPI
+from pydantic import BaseModel
+from llama_cpp import Llama
+from huggingface_hub import hf_hub_download
+
+app = FastAPI()
+
+# Downloads your specific model
+model_path = hf_hub_download(
+    repo_id="devray11/Aevis-Medical-SLM",
+    filename="DeepSeek-R1-Distill-Llama-8B.Q4_K_M.gguf"
+)
+
+# Initializes the AI engine
+llm = Llama(model_path=model_path, n_ctx=2048)
+
+class Query(BaseModel):
+    prompt: str
+
+@app.post("/generate")
+async def generate(query: Query):
+    # Formats the prompt exactly like your training data
+    fmt_prompt = f"### Instruction:\n{query.prompt}\n\n### Response:\n"
+    output = llm(fmt_prompt, max_tokens=512, stop=["###"])
+    return {"response": output["choices"][0]["text"]}
+
+@app.get("/")
+def health():
+    return {"status": "Aevis API is Online"}