Chat completions

POST /<lane>/v1/chat/completions — the OpenAI Chat Completions wire protocol. Used by Cursor, Cline/Roo, OpenCode (openai-compatible), and the OpenAI SDK.

Request

curl "https://api.opengateway.one/oss/v1/chat/completions" \
  -H "Authorization: Bearer $OPENGATEWAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-turbo-hub-qwen3-coder",
    "messages": [
      { "role": "system", "content": "You are a precise coding assistant." },
      { "role": "user", "content": "Write a TypeScript debounce helper." }
    ],
    "reasoning_effort": "medium"
  }'

Response

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "model": "claude-turbo-hub-qwen3-coder",
  "choices": [
    {
      "index": 0,
      "message": { "role": "assistant", "content": "export function debounce..." },
      "finish_reason": "stop"
    }
  ],
  "usage": { "prompt_tokens": 24, "completion_tokens": 88, "total_tokens": 112 }
}

Streaming

Set stream: true to receive chat.completion.chunk SSE events.

OpenAI (TS)
OpenAI (Python)

const stream = await client.chat.completions.create({
  model: 'claude-turbo-hub-qwen3-coder',
  messages: [{ role: 'user', content: 'Stream a haiku about the edge.' }],
  stream: true,
});
for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? '');
}

stream = client.chat.completions.create(
    model="claude-turbo-hub-qwen3-coder",
    messages=[{"role": "user", "content": "Stream a haiku about the edge."}],
    stream=True,
)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")

Notes

reasoning_effort accepts none|minimal|low|medium|high|xhigh — see Effort levels.
Tool calls (tools/tool_calls) are passed through and preserved across streaming.
On /hf, use any org/model[:policy] id; on /oss, use a curated alias.