OpenAI API Automation: Complete Production Reference

import openai, os, json
client = openai.OpenAI(api_key=os.environ["OPENAI_API_KEY"])

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[
        {"role": "system", "content": "Your system prompt."},
        {"role": "user", "content": "The input to process."}
    ],
    temperature=0.2,
    max_tokens=300,
    response_format={"type": "json_object"}
)
result = json.loads(response.choices[0].message.content)

r = client.embeddings.create(
    model="text-embedding-3-small",
    input=["Text to convert to vector"]
)
vector = r.data[0].embedding  # List of 1536 floats

with open("meeting.mp3", "rb") as f:
    result = client.audio.transcriptions.create(
        model="whisper-1", file=f, language="en"
    )
transcript = result.text  # Cost: $0.006 per minute of audio

import base64
with open("invoice.jpg","rb") as f:
    img = base64.b64encode(f.read()).decode()
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role":"user","content":[
        {"type":"text","text":"Extract invoice data as JSON."},
        {"type":"image_url","image_url":{"url":f"data:image/jpeg;base64,{img}"}}
    ]}],
    response_format={"type":"json_object"}
)

import time
from openai import RateLimitError, APIStatusError, APIConnectionError

def api_with_retry(messages, model="gpt-4o-mini", max_retries=3):
    for attempt in range(max_retries):
        try:
            r = client.chat.completions.create(
                model=model, messages=messages, temperature=0.2,
                response_format={"type":"json_object"}
            )
            return r.choices[0].message.content
        except RateLimitError as e:
            if "insufficient_quota" in str(e): raise  # Budget gone — don't retry
            time.sleep(2 ** attempt)
        except APIStatusError as e:
            if e.status_code == 400: raise  # Code bug — don't retry
            time.sleep(2 ** attempt * 2)
        except APIConnectionError:
            time.sleep(2 ** attempt * 5)
    raise Exception(f"API failed after {max_retries} attempts")

response = client.chat.completions.create(...)
usage = response.usage

# Prices Nov 2024 — verify current at platform.openai.com/pricing
PRICES = {
    "gpt-4o-mini-2024-07-18": {"in": 0.15/1e6, "out": 0.60/1e6},
    "gpt-4o-2024-08-06":      {"in": 2.50/1e6, "out": 10.0/1e6},
}
m = response.model
cost = usage.prompt_tokens * PRICES.get(m, PRICES["gpt-4o-mini-2024-07-18"])["in"] +        usage.completion_tokens * PRICES.get(m, PRICES["gpt-4o-mini-2024-07-18"])["out"]
print(f"  Tokens: {usage.total_tokens} | Cost: ${cost:.5f}")

import json

# Build JSONL request file
requests_data = []
for item in items_to_process:
    requests_data.append(json.dumps({
        "custom_id": item["id"],
        "method": "POST",
        "url": "/v1/chat/completions",
        "body": {
            "model": "gpt-4o-mini",
            "messages": [
                {"role": "system", "content": SYSTEM_PROMPT},
                {"role": "user", "content": item["content"]}
            ],
            "response_format": {"type": "json_object"}
        }
    }))

with open("batch.jsonl", "w") as f:
    f.write("
".join(requests_data))

# Submit
with open("batch.jsonl", "rb") as f:
    bf = client.files.create(file=f, purpose="batch")
batch = client.batches.create(
    input_file_id=bf.id, endpoint="/v1/chat/completions", completion_window="24h"
)
print(f"Batch submitted: {batch.id} — results within 24 hours")