Other Clients & Protocols

OpenAI compatible

Chat Completions and Responses endpoints, for the OpenAI SDK and any OpenAI-compatible client.

For the official OpenAI SDK and OpenAI-compatible clients. The key’s group must support OpenAI. Other vendors’ models, including Claude, can use this protocol where supported.

Endpoint and authentication

YOUR_API_KEY in the examples is a placeholder and will not work as copied. Sign in to pick one of your keys and fill it into every example on this page.

SDK base_url

https://api.keyvia.cc/v1

Include /v1

Chat Completions

POST https://api.keyvia.cc/v1/chat/completions

Responses

POST https://api.keyvia.cc/v1/responses

Model list

GET https://api.keyvia.cc/v1/models

Use the authentication header Authorization: Bearer YOUR_API_KEY. The example model comes from the catalog; replace it with another available model if needed.

Chat Completions

Send messages. The reply is in choices[0].message.content.

curl

curl https://api.keyvia.cc/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5",
    "messages": [{ "role": "user", "content": "Reply only: Hello" }]
  }'

Python · pip install openai

from openai import OpenAI

client = OpenAI(api_key="YOUR_API_KEY", base_url="https://api.keyvia.cc/v1")
resp = client.chat.completions.create(
    model="gpt-5",
    messages=[{"role": "user", "content": "Reply only: Hello"}],
)
print(resp.choices[0].message.content)

Node.js · npm install openai

import OpenAI from 'openai'

const client = new OpenAI({ apiKey: 'YOUR_API_KEY', baseURL: 'https://api.keyvia.cc/v1' })
const resp = await client.chat.completions.create({
  model: 'gpt-5',
  messages: [{ role: 'user', content: 'Reply only: Hello' }],
})
console.log(resp.choices[0].message.content)

Responses

This is the protocol used by Codex. Send input. The reply is in output entries with type output_text. The SDK provides output_text convenience property.

curl

curl https://api.keyvia.cc/v1/responses \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "gpt-5", "input": "Reply only: Hello" }'

Python

from openai import OpenAI

client = OpenAI(api_key="YOUR_API_KEY", base_url="https://api.keyvia.cc/v1")
resp = client.responses.create(model="gpt-5", input="Reply only: Hello")
print(resp.output_text)

Streaming

Add this to the request body: "stream": true to receive SSE events in the upstream format. In the SDK, set stream=True and iterate over the chunks.

Python

from openai import OpenAI

client = OpenAI(api_key="YOUR_API_KEY", base_url="https://api.keyvia.cc/v1")
stream = client.chat.completions.create(
    model="gpt-5",
    messages=[{"role": "user", "content": "Introduce yourself in one sentence"}],
    stream=True,
)
for chunk in stream:
    delta = chunk.choices[0].delta.content
    if delta:
        print(delta, end="", flush=True)

See available models in the Model catalog. Models and parameters depend on the gateway’s supported routes. For failures, see Error codes.