OpenAI Python SDK 3.37.0: our test log

The plain OpenAI Python SDK (v3.37.0) with a hand-written tool loop passed 3 of 3 trials against a local model; it is our control for every framework test.

Result at a glance

What we found

• 3 of 3 task trials passed: the model's tool call reached the framework, the function ran with 17 and 25, and the final answer contained 42.
• Cold install took 1.1 s with uv and 7.2 s with pip, and left 26 MB in 14 packages (uv).
• This is the control: no agent framework at all, just the client library and a hand-written loop (about 15 lines). It is the fastest and smallest install in the study, which is the cost every framework below adds to.

What we ran, step by step

# Install with uv (cold, no cache)
uv venv -q -p 3.13 work/py/openai-python && VIRTUAL_ENV=work/py/openai-python uv pip install --no-cache -q openai

# Install with pip (cold, no cache)
uv venv -q --seed -p 3.13 work/pip/openai-python && work/pip/openai-python/bin/pip install --no-cache-dir -q openai

# Task command (all trials)
OTEL_SDK_DISABLED=true work/py/openai-python/bin/python tasks/openai_python.py

Raw output (excerpts)

from common import *
import json
from openai import OpenAI
c = OpenAI(base_url=BASE, api_key=KEY)
tools = [{"type":"function","function":{"name":"add_numbers","description":"Add two integers and return the sum.","parameters":{"type":"object","properties":{"a":{"type":"integer"},"b":{"type":"integer"}},"required":["a","b"]}}}]
msgs = [{"role":"user","content":TASK}]
final = ""
for _ in range(4):
    r = c.chat.completions.create(model=MODEL, messages=msgs, tools=tools)
    m = r.choices[0].message
    if m.tool_calls:
        msgs.append(m.model_dump(exclude_none=True))
        for tc in m.tool_calls:
            args = json.loads(tc.function.arguments)
            msgs.append({"role":"tool","tool_call_id":tc.id,"content":str(add_numbers(**args))})
    else:
        final = m.content; break
report(final)

import json, os
BASE = "http://127.0.0.1:11434/v1"
MODEL = os.environ.get("MODEL", "qwen2.5-3b")
KEY = "ollama"
TASK = "Use the add_numbers tool to compute 17 + 25, then tell me the result."
CALLS = []
def add_numbers(a: int, b: int) -> int:
    """Add two integers and return the sum."""
    CALLS.append((a, b))
    return int(a) + int(b)
def report(final):
    ok = any({int(x), int(y)} == {17, 25} for x, y in CALLS) and "42" in str(final)
    print("RESULT_JSON " + json.dumps({"calls": CALLS, "final": str(final)[:300], "ok": ok}))

RESULT_JSON {"calls": [[17, 25]], "final": "The result of adding 17 and 25 is 42.", "ok": true}

Gotchas and things that broke

• The loop is ours, so this test measures the local model and the Ollama tool-calling layer, not an agent framework.

What we did not test

• Any vendor-hosted model: everything ran on a local 3B model, so tool-call reliability with frontier models is not measured here.
• Multi-agent flows, memory, long conversations, streaming, tracing and cloud deployment.
• Anything beyond one tool and one question; three trials is a smoke test, not a benchmark.

Reproduce it

Versions change. These commands install whatever is latest on the day you run them unless a version is pinned, so expect different results later. The version we measured is in the table above.

# 1. Serve any local model that supports tool calling behind an OpenAI-compatible endpoint, registered as qwen2.5-3b:
#    http://127.0.0.1:11434/v1  (we used Ollama with a Qwen2.5-3B-Instruct Q4_K_M GGUF)
uv venv -p 3.13 .venv && VIRTUAL_ENV=.venv uv pip install openai
# 2. Save common.py (above) and the task script next to each other, then:
.venv/bin/python openai_python.py

How to read this test

Every number on this page was measured on this one machine, which also ran other jobs, so timings are indicative and vary run to run. A pass means our small task worked once or as stated, a fail means it did not work as we ran it; neither says how the tool behaves with a vendor's frontier model, on a GPU, or at scale. Where a tool could not run without an account or a GPU, we say so instead of guessing.

See the testing rig for the hardware, the local model and the rules we follow, and the test hub for every other test.

Related tests: agent frameworks and sdks

• Pydantic AI: Pydantic AI 2.61.0 passed our one-tool task on a local 3B model: 3 of 3 trials; cold uv install 3.7 s.
• Semantic Kernel: Semantic Kernel 1.45.0 returned an empty reply with our standard prompt (0 of 3) but worked once the prompt used its plugin-prefixed tool name (3 of 3); a naming trap with small models.
• smolagents: smolagents 1.26.0 passed our one-tool task on a local 3B model: 6 of 6 trials; cold uv install 1.5 s.
• Vercel AI SDK: Vercel AI SDK passed our one-tool task 3 of 3 on Node 20 (ai 6.0.315) and 3 of 3 on Node 24 (ai 7.0.153); the install is 26 to 27 MB.
• Agno: Agno 3.1.10 passed our one-tool task on a local 3B model: 3 of 3 trials; cold uv install 1.8 s.
• AutoGen: AutoGen 0.7.5 passed our one-tool task on a local 3B model: 3 of 3 trials; cold uv install 1.9 s.

All test logs · How we test

FAQ

Sources

• openai-python repository (fetched 2026-10-25)

Related

• All test logs
• Methodology and testing rig
• Study: install size and first-task results across 16 frameworks