# GPT-6 Sol — API integration preview (reAPI)

> API not available on reAPI yet. This is a proposed request format, not a tested integration. The playground only builds JSON locally; it does not send requests or generate model responses.

## Availability

- Status: preview (prelaunch; not live; not yet callable on reAPI)
- Planned model id: `gpt-6-sol`
- Planned protocol: OpenAI-compatible Chat Completions.
- Planned endpoint: `POST https://reapi.ai/api/v1/chat/completions`
- Pricing: not announced. No live price or billing estimate is available for this model.

## Planned request parameters

This is a proposed integration format, not a live request schema. Required fields, options and defaults below describe the draft only. Official model limits do not confirm reAPI support; parameter behavior has not been verified through reAPI.

- `model` (string, required in this draft) — Proposed model ID: gpt-6-sol. Not currently served by reAPI.
- `messages` (array, required in this draft) — Chat history. This preview builds a system message (optional) and a user text message. Image inputs require separate integration verification.
- `max_completion_tokens` (integer, optional in this draft) — Preview default: 4096. Official output limit: 128000 tokens, including reasoning tokens. The reAPI limit is not yet confirmed.
- `reasoning_effort` (string, optional in this draft) — none, low, medium (default), high, xhigh, max. Official Chat Completions function calling requires none; reAPI support is pending.
- `stream` (boolean, optional in this draft) — Proposed streaming response option. The preview defaults to true; no stream is opened.

## Draft request example — not live

For offline integration planning only. This JSON does not demonstrate a successful request or a verified model response.

```json
{
  "model": "gpt-6-sol",
  "messages": [
    {
      "role": "user",
      "content": "Review this migration plan for risks and propose a test checklist."
    }
  ],
  "max_completion_tokens": 4096,
  "reasoning_effort": "medium",
  "stream": true
}
```

## Draft cURL — do not run before launch

```bash
# Draft only — GPT-6 Sol is not available on reAPI yet. Do not run.
curl https://reapi.ai/api/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  --no-buffer \
  -d '{
  "model": "gpt-6-sol",
  "messages": [
    {
      "role": "user",
      "content": "Review this migration plan for risks and propose a test checklist."
    }
  ],
  "max_completion_tokens": 4096,
  "reasoning_effort": "medium",
  "stream": true
}'
```

The endpoint, authentication, response and streaming behavior must be confirmed when reAPI integration is available. These examples do not enable API access.

## Related documentation

- Draft API reference (Markdown): https://reapi.ai/docs/gpt-6-sol.md
- Draft API reference (HTML): https://reapi.ai/docs/gpt-6-sol
- Model preview and launch status: https://reapi.ai/models/gpt-6-sol
- Official model documentation (vendor capabilities, not reAPI availability): https://developers.openai.com/api/docs/models/gpt-6-sol
- Platform index: https://reapi.ai/llms.txt
- Live API specification (this preview does not register a live model): https://reapi.ai/openapi.json