-
Notifications
You must be signed in to change notification settings - Fork 271
Expand file tree
/
Copy pathchat_async.py
More file actions
85 lines (70 loc) · 2.7 KB
/
chat_async.py
File metadata and controls
85 lines (70 loc) · 2.7 KB
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
import asyncio
import os
import azure.identity.aio
import openai
from dotenv import load_dotenv
# Configura el cliente de OpenAI para usar la API de Azure, OpenAI.com u Ollama
load_dotenv(override=True)
API_HOST = os.getenv("API_HOST", "github")
azure_credential = None # Guarda la Azure Credential para poder cerrarla correctamente
if API_HOST == "azure":
azure_credential = azure.identity.aio.DefaultAzureCredential()
token_provider = azure.identity.aio.get_bearer_token_provider(
azure_credential, "https://cognitiveservices.azure.com/.default"
)
client = openai.AsyncOpenAI(
base_url=os.environ["AZURE_OPENAI_ENDPOINT"],
api_key=token_provider,
)
MODEL_NAME = os.environ["AZURE_OPENAI_CHAT_DEPLOYMENT"]
elif API_HOST == "ollama":
client = openai.AsyncOpenAI(base_url=os.environ["OLLAMA_ENDPOINT"], api_key="nokeyneeded")
MODEL_NAME = os.environ["OLLAMA_MODEL"]
elif API_HOST == "github":
client = openai.AsyncOpenAI(base_url="https://models.github.ai/inference", api_key=os.environ["GITHUB_TOKEN"])
MODEL_NAME = os.getenv("GITHUB_MODEL", "openai/gpt-4o")
else:
client = openai.AsyncOpenAI(api_key=os.environ["OPENAI_KEY"])
MODEL_NAME = os.environ["OPENAI_MODEL"]
async def generate_response(location):
print("Generando respuesta para", location)
response = await client.chat.completions.create(
model=MODEL_NAME,
messages=[
{"role": "system", "content": "Eres un asistente útil."},
{
"role": "user",
"content": (
f"Nombra un solo lugar que debería visitar en mi viaje a {location} y descríbelo en una oración"
),
},
],
temperature=0.7,
)
print("Obtuve respuesta para ", location)
return response.choices[0].message.content
async def single() -> None:
"""Ejecuta un único ejemplo."""
print(await generate_response("Tokio"))
async def multiple() -> None:
"""Ejecuta varias solicitudes concurrentes."""
answers = await asyncio.gather(
generate_response("Tokio"),
generate_response("Berkeley"),
generate_response("Singapur"),
)
for answer in answers:
print(answer, "\n")
async def close_clients() -> None:
"""Cierra el cliente OpenAI y la credencial de Azure (si existe)."""
await client.close()
if azure_credential is not None:
await azure_credential.close()
async def main():
"""Punto de entrada que garantiza la liberación de recursos."""
try:
await single() # Usa await multiple() si quieres ejecutar de forma concurrente.
finally:
await close_clients()
if __name__ == "__main__":
asyncio.run(main())