Native Invocation (Responses API)

In addition to the classic Chat Completions, 4ALL API also supports OpenAI’s new native interface Responses API (/v1/responses). Some new tools and clients (such as Codex CLI) use this protocol by default.

Basic Invocation

curl https://api.4allapi.com/v1/responses \
  -H "Authorization: Bearer $AIPROXY_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.5",
    "input": "Explain what an API gateway is in one sentence"
  }'

input can be either a string or an array of messages with roles:

{
  "model": "gpt-5.5",
  "input": [
    { "role": "system", "content": "You are a concise Chinese assistant" },
    { "role": "user", "content": "What is an API gateway?" }
  ]
}

Differences from Chat Completions

Point	Chat Completions	Responses
Input field	messages	input
Output location	choices[0].message.content	message items in the output array
Suitable for	Most clients and SDKs	OpenAI’s new toolchain

Recommendations

• For existing services, Chat Completions is sufficient, no migration needed;
• When a client strictly requires the Responses protocol (such as Codex CLI), point it directly to https://api.4allapi.com/v1, and the protocol will be available automatically;
• Capabilities such as streaming and function calling are supported by both protocols, and the billing rules are the same.