MiniMax 模型 | API 文件

# MiniMax Model

Call MiniMax M3 through the shared Chat Completions endpoint.

POST https://api.apihop.ai/v1/chat-completions

MiniMax M3 is available through the Common Chat Completions endpoint, with both non-streaming and SSE streaming text responses.

```text
POST https://api.apihop.ai/v1/chat-completions
```

## Available Model

| `model` value | Type | Non-streaming | SSE streaming |
| --- | --- | --- | --- |
| `minimax-m3` | Text chat | Yes | Yes |

## Authentication

```http
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json
```

## Non-streaming Request

```json
{
  "model": "minimax-m3",
  "messages": [
    {
      "role": "system",
      "content": "You are a code review assistant."
    },
    {
      "role": "user",
      "content": "List five risk areas to review in a payment webhook implementation."
    }
  ],
  "thinking": {
    "type": "disabled"
  },
  "stream": false
}
```

## Streaming Request

```json
{
  "model": "minimax-m3",
  "messages": [
    {
      "role": "user",
      "content": "Analyze a duplicate queue-consumption issue step by step."
    }
  ],
  "stream": true
}
```

This endpoint returns one complete JSON response by default. Set `stream: true` explicitly when you need SSE streaming.

## Request Fields

### `model` (string, required)

Use `minimax-m3`.

### `messages` (object[], required)

OpenAI Chat Completions style messages. Each message contains `role` and `content`.

### `stream` (boolean, Default: false)

`true` returns SSE; `false` returns one complete JSON response.

### `max_tokens` (integer)

Optional maximum output tokens, as a positive integer. Omit this field to use the upstream default for the selected model. Depending on the model, the budget may include reasoning tokens as well as final text. A small budget can truncate the answer or be exhausted before final text; increasing it may increase cost.

### `max_completion_tokens` (integer)

Optional compatibility alternative to `max_tokens`, with the same budget considerations. The selected model determines support; omit both fields to use the upstream default.

### `thinking` (object)

MiniMax thinking control. Use `{ "type": "disabled" }` for short, direct answers that do not need hidden reasoning. Omit the field to keep the model's default thinking behavior.

See [Common Chat Completions](https://apihop.ai/en/docs/api-reference/common-chat-completions) for shared fields such as `temperature`, `top_p`, `frequency_penalty`, `presence_penalty`, `stop`, and `n`. `minimax-m3` accepts these standard fields, but not every parameter necessarily produces a measurable semantic change.

## Success Response

```json
{
  "id": "chatcmpl_...",
  "object": "chat.completion",
  "created": 1787200000,
  "model": "minimax-m3",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Start by checking idempotency keys, acknowledgements, and retry policy..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 29,
    "completion_tokens": 104,
    "total_tokens": 133
  }
}
```

`usage` may include optional cached-input or reasoning-token details. Continue to base accounting and telemetry on the standard input, output, and total token fields.

## Usage Notes

- The current public contract is Chat Completions, including non-streaming and SSE streaming.
- `minimax-m3` is not currently published as an OpenAI Responses or Claude Messages model.
- Use `minimax-m3` exactly.
- Use `GET https://api.apihop.ai/v1/models` for the models visible to the current API Key, and use the console for current pricing.

## Related Docs

- [Common Chat Completions](https://apihop.ai/en/docs/api-reference/common-chat-completions)
- [Model Catalog](https://apihop.ai/en/docs/api-reference/models)

MiniMax M3 可透過共用 Chat Completions 端點使用，支援非串流及 SSE 串流文字回應。

text

POST https://api.apihop.ai/v1/chat-completions

可用的模型

`model` 值	類型	非串流	SSE 串流
`minimax-m3`	文字對話	支援	支援

驗證

http

Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

非串流請求

json

{
  "model": "minimax-m3",
  "messages": [
    {
      "role": "system",
      "content": "You are a code review assistant."
    },
    {
      "role": "user",
      "content": "List five risk areas to review in a payment webhook implementation."
    }
  ],
  "thinking": {
    "type": "disabled"
  },
  "stream": false
}

串流請求

json

{
  "model": "minimax-m3",
  "messages": [
    {
      "role": "user",
      "content": "Analyze a duplicate queue-consumption issue step by step."
    }
  ],
  "stream": true
}

此端點預設回傳一個完整 JSON 回應。需要 SSE 串流時，請明確設定 stream: true。

請求欄位

modelstring必填

使用 minimax-m3。

messagesobject[]必填

OpenAI Chat Completions 風格的訊息，每則訊息包含 role 與 content。

streamboolean預設: false

true 回傳 SSE；false 回傳一個完整 JSON 回應。

max_tokensinteger

選填的輸出 Token 上限，須為正整數。省略時使用所選模型的上游預設值。依模型而定，此額度可能同時包含推理 Token 與最終文字。過小的額度可能截斷回答，或在產生最終文字前耗盡；提高額度可能增加費用。

max_completion_tokensinteger

max_tokens 的選填相容替代欄位，具有相同額度注意事項。支援情況由所選模型決定；省略兩個欄位可使用上游預設值。

thinkingobject

MiniMax 推理控制。對不需要隱藏推理的簡短直接回答，使用 { "type": "disabled" }。省略欄位可保留模型預設推理行為。

關於 temperature、top_p、frequency_penalty、presence_penalty、stop 與 n 等共用欄位，請參閱共用 Chat Completions。minimax-m3 接受這些標準欄位，但不是每個參數都一定產生可量測的語意變化。

成功回應

json

{
  "id": "chatcmpl_...",
  "object": "chat.completion",
  "created": 1787200000,
  "model": "minimax-m3",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Start by checking idempotency keys, acknowledgements, and retry policy..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 29,
    "completion_tokens": 104,
    "total_tokens": 133
  }
}

usage 可能包含選填的快取輸入或推理 Token 明細。核算與遙測仍應以標準輸入、輸出及總 Token 欄位為依據。

使用注意事項

目前公開協定為 Chat Completions，包含非串流與 SSE 串流。
minimax-m3 目前未以 OpenAI Responses 或 Claude Messages 模型發布。
請精確使用 minimax-m3。
以 GET https://api.apihop.ai/v1/models 查看目前金鑰可見的模型，並在控制台查看現行價格。