What is the Chat Completions API and when should you still use it?

The Chat Completions API (/v1/chat/completions) is OpenAI's original and most widely adopted API, introduced with GPT-3.5 and GPT-4. It uses an array of messages with role (system, user, assistant) and content fields to generate responses.

from openai import OpenAI
client = OpenAI()

# Classic Chat Completions call
response = client.chat.completions.create(
    model="gpt-5.5",
    messages=[
        {"role": "system", "content": "You are an expert Python developer."},
        {"role": "user", "content": "Explain asyncio.gather() with an example."},
    ],
    temperature=0.2,
    max_tokens=1024,
)
print(response.choices[0].message.content)

When to continue using Chat Completions:

• Your application does NOT need built-in tools (web search, file search, computer use)
• You have extensive existing integrations built on Chat Completions and migration is not yet justified
• You need maximum compatibility with third-party libraries and frameworks
• Simple, single-turn completions without stateful context

Important note for reasoning models: starting with GPT-5.4, tool calling is NOT supported in Chat Completions with reasoning: none. For tool use with reasoning models, migrate to the Responses API.

OpenAI has committed to continuing to release new models to Chat Completions as long as those models' capabilities don't depend on built-in tools or multiple model calls.