Spaces:
Sleeping
Sleeping
File size: 1,219 Bytes
33ecdc5 d655b62 2ec91ed d655b62 bb9adf1 d655b62 2ec91ed d655b62 dcb6f6f d655b62 dcb6f6f d655b62 dcb6f6f |
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 |
import os
import requests
# Replace this with your actual endpoint URL
API_URL = "https://lgj704z9p0j2vf79.us-east4.gcp.endpoints.huggingface.cloud"
HF_ENDPOINT_TOKEN = os.environ.get("HF_ENDPOINT_TOKEN")
headers = {
"Authorization": f"Bearer {HF_ENDPOINT_TOKEN}",
"Content-Type": "application/json"
}
def call_model(prompt: str) -> str:
response = requests.post(
f"{API_URL}/generate", # <-- use /generate for HF endpoints
headers=headers,
json={
"inputs": prompt,
"parameters": {
"max_new_tokens": 2048,
"temperature": 0.3,
"do_sample": False
}
}
)
if response.status_code != 200:
raise RuntimeError(f"Inference error: {response.status_code} - {response.text}")
result = response.json()
# Handle variations in response format
if isinstance(result, dict) and "generated_text" in result:
return result["generated_text"]
elif isinstance(result, list) and "generated_text" in result[0]:
return result[0]["generated_text"]
elif "text" in result:
return result["text"]
else:
return "⚠️ No output generated."
|