A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 4, 15:00 UTC
[Bugfix] Fix broken example opeanai client (#41088)
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Isotr0py committed
Apr 28, 2026 at 12:43 UTC
c2e88a281c53059d023a2aee43217a7379509a4a
3 files changed
+6
-9
examples/generate/multimodal/openai_chat_completion_client_for_multimodal.py
+1
-2
@@ -25,7 +25,6 @@ import os
25
import pybase64 as base64
26
import requests
27
from openai import OpenAI
28
-from utils import get_first_model
28
29
from vllm.utils.argparse_utils import FlexibleArgumentParser
30
@@ -407,7 +406,7 @@ def parse_args():
406
407
def main(args) -> None:
408
chat_type = args.chat_type
410
- model = get_first_model(client)
409
+ model = client.models.list().data[0].id
410
example_function_map[chat_type](model, args.max_completion_tokens)
411
412
examples/tool_calling/openai_responses_client_with_mcp_tools.py
+4
-5
@@ -19,7 +19,6 @@ Environment variables:
19
"""
20
21
from openai import OpenAI
22
-from utils import get_first_model
22
23
24
def example_no_filter():
@@ -30,7 +29,7 @@ def example_no_filter():
29
30
base_url = "http://0.0.0.0:8000/v1"
31
client = OpenAI(base_url=base_url, api_key="empty")
33
- model = get_first_model(client)
32
+ model = client.models.list().data[0].id
33
34
response = client.responses.create(
35
model=model,
@@ -59,7 +58,7 @@ def example_wildcard():
58
59
base_url = "http://0.0.0.0:8000/v1"
60
client = OpenAI(base_url=base_url, api_key="empty")
62
- model = get_first_model(client)
61
+ model = client.models.list().data[0].id
62
63
response = client.responses.create(
64
model=model,
@@ -95,7 +94,7 @@ def example_specific_tools():
94
95
base_url = "http://0.0.0.0:8000/v1"
96
client = OpenAI(base_url=base_url, api_key="empty")
98
- model = get_first_model(client)
97
+ model = client.models.list().data[0].id
98
99
response = client.responses.create(
100
model=model,
@@ -126,7 +125,7 @@ def example_object_format():
125
126
base_url = "http://0.0.0.0:8000/v1"
127
client = OpenAI(base_url=base_url, api_key="empty")
129
- model = get_first_model(client)
128
+ model = client.models.list().data[0].id
129
130
response = client.responses.create(
131
model=model,
examples/tool_calling/openai_responses_client_with_tools.py
+1
-2
@@ -14,7 +14,6 @@ vllm serve Qwen/Qwen3-1.7B --reasoning-parser qwen3 \
14
import json
15
16
from openai import OpenAI
17
-from utils import get_first_model
17
18
19
def get_weather(latitude: float, longitude: float) -> str:
@@ -51,7 +50,7 @@ input_messages = [
50
def main():
51
base_url = "http://0.0.0.0:8000/v1"
52
client = OpenAI(base_url=base_url, api_key="empty")
54
- model = get_first_model(client)
53
+ model = client.models.list().data[0].id
54
response = client.responses.create(
55
model=model, input=input_messages, tools=tools, tool_choice="required"
56
)