File size: 1,156 Bytes
9f4e457 ac58288 9f4e457 27d8de0 9f4e457 27d8de0 2d88192 27d8de0 9f4e457 27d8de0 9f4e457 27d8de0 9f4e457 27d8de0 9f4e457 ac58288 27d8de0 9f4e457 27d8de0 bdf7da6 9f4e457 27d8de0 9f4e457 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 |
from fastapi import FastAPI, Form
from fastapi.middleware.cors import CORSMiddleware
from pydantic import BaseModel
from ctransformers import AutoModelForCausalLM
#Model loading
llm = AutoModelForCausalLM.from_pretrained("zephyr-7b-beta.Q4_K_S.gguf",
model_type='mistral',
max_new_tokens = 256,
threads = 3,
)
#Pydantic object
class validation(BaseModel):
prompt: str
#Fast API
app = FastAPI()
# Set up CORS
origins = [
"http://localhost", # Replace with the address of your Flutter web app
"http://localhost:55345", # Add the port used by your Flutter web app
]
app.add_middleware(
CORSMiddleware,
allow_origins=origins,
allow_credentials=True,
allow_methods=["*"],
allow_headers=["*"],
)
#Zephyr completion
@app.post("/llm_on_cpu")
async def stream(item: validation):
system_prompt = 'Your name is Maff, a mentor who teaches people how to make money. Follow these guideline- Keep your sentences short.'
E_INST = "</s>"
user, assistant = "<|user|>", "<|assistant|>"
prompt = f"{system_prompt}{E_INST}\n{user}\n{item.prompt.strip()}{E_INST}\n{assistant}\n"
return llm(prompt)
|