Gemini 模型 | API 文件

# Gemini Models

Call Gemini 3.6 Flash, 3.7 Flash, 3.8 Flash, 3.5 Flash-Lite, and 4 Argon through the shared Chat Completions endpoint.

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

Gemini models use the Common Chat Completions endpoint. This page records the supported `model` values and text chat contract.

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

## Available Models

| `model` value | Non-streaming JSON | SSE streaming |
| --- | --- | --- |
| `gemini-3.6-flash` | Yes | Yes |
| `gemini-3.7-flash` | Yes | Yes |
| `gemini-3.8-flash` | Yes | Yes |
| `gemini-3.5-flash-lite` | Yes | Yes |
| `gemini-4-argon` | Yes | Yes |

`gemini-4-argon` requires deployment and enablement. Its inclusion in this reference does not confirm production availability.

Model names distinguish product tiers; they are not latency, throughput, or quality SLAs. Compare models with your real workload. Check `GET https://api.apihop.ai/v1/models` for availability to your API Key before making a request.

## Authentication

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

## Non-streaming Request

```json
{
  "model": "gemini-3.7-flash",
  "messages": [
    {
      "role": "system",
      "content": "You are a concise and accurate technical assistant."
    },
    {
      "role": "user",
      "content": "Explain idempotent APIs in five sentences."
    }
  ],
  "stream": false
}
```

## Streaming Request

```json
{
  "model": "gemini-3.6-flash",
  "messages": [
    {
      "role": "user",
      "content": "Show a TypeScript API error-handling example."
    }
  ],
  "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 `gemini-3.6-flash`, `gemini-3.7-flash`, `gemini-3.8-flash`, `gemini-3.5-flash-lite`, or `gemini-4-argon`.

### `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.

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`. Support and behavior of optional fields depend on the selected model; this page does not guarantee every shared field.

## Success Response

```json
{
  "id": "chatcmpl_...",
  "object": "chat.completion",
  "created": 1787200000,
  "model": "gemini-3.7-flash",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "An idempotent API produces the same effective result when..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 26,
    "completion_tokens": 118,
    "total_tokens": 144
  }
}
```

`usage` may include optional reasoning-token details; clients should tolerate them being absent or changing by model.

## Usage Notes

- The request body's `model` must be one of the public model IDs listed in the Available Models table.
- Product names, display names, and other aliases are not accepted in place of a public model ID.
- This page applies only to Chat Completions requests sent to

`POST https://api.apihop.ai/v1/chat-completions`, not OpenAI Responses or Claude Messages.

- 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)

Gemini 模型使用共用 Chat Completions 端點。本頁記錄支援的 model 值與文字對話協定。

text

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

可用模型

`model` 值	非串流 JSON	SSE 串流
`gemini-3.6-flash`	支援	支援
`gemini-3.7-flash`	支援	支援
`gemini-3.8-flash`	支援	支援
`gemini-3.5-flash-lite`	支援	支援
`gemini-4-argon`	支援	支援

gemini-4-argon 需要部署並啟用。列入本參考文件不代表正式環境已可用。

模型名稱區分產品層級，不是延遲、吞吐量或品質 SLA。請以實際工作負載比較模型；發出請求前，用 GET https://api.apihop.ai/v1/models 確認金鑰可用性。

驗證

http

Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

非串流請求

json

{
  "model": "gemini-3.7-flash",
  "messages": [
    {
      "role": "system",
      "content": "You are a concise and accurate technical assistant."
    },
    {
      "role": "user",
      "content": "Explain idempotent APIs in five sentences."
    }
  ],
  "stream": false
}

串流請求

json

{
  "model": "gemini-3.6-flash",
  "messages": [
    {
      "role": "user",
      "content": "Show a TypeScript API error-handling example."
    }
  ],
  "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 gemini-3.6-flash 或 gemini-3.7-flash 或 gemini-3.8-flash 或 gemini-3.5-flash-lite 或 gemini-4-argon。

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。

關於 temperature、top_p、frequency_penalty、presence_penalty、stop 與 n 等共用欄位，請參閱共用 Chat Completions。選填欄位的支援與行為依模型而定；本頁不保證所有共用欄位都可用。

成功回應

json

{
  "id": "chatcmpl_...",
  "object": "chat.completion",
  "created": 1787200000,
  "model": "gemini-3.7-flash",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "An idempotent API produces the same effective result when..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 26,
    "completion_tokens": 118,
    "total_tokens": 144
  }
}

usage 可能包含選填的推理 Token 明細；用戶端應容許其缺少或因模型而變化。

使用注意事項

主體的 model 必須是可用模型表中的公開 ID 之一。
不接受產品名稱、顯示名稱或其他別名取代公開模型 ID。
本頁僅適用於傳至 POST https://api.apihop.ai/v1/chat-completions 的 Chat Completions 請求，不適用 OpenAI Responses 或 Claude Messages。
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。