# 1. 服务部署
python -m vlm.entrypoints.openai_api_server --model /root/models/Meta-llama-3-8B --dtype auto --api-key 123456
# 2. 服务测试(vlm_completion_test.py)
from openai import OpenAI
client = OpenAI(
base_url="http://localhost:8000/v1",
api_key="123456",
)
print("服务连接成功")
completion = client.completions.create(
model="/root/models/Meta-llama-3-8B",
prompt="San Francisco is a",
max_tokens=128,
)
print("# San Francisco is : ")
print("Completion result: ", completion)
03-19
1158
![](https://csdnimg.cn/release/blogv2/dist/pc/img/readCountWhite.png)
05-17
06-19
“相关推荐”对你有帮助么?
-
非常没帮助
-
没帮助
-
一般
-
有帮助
-
非常有帮助
提交