> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tokenlab.sh/llms.txt
> Use this file to discover all available pages before exploring further.

# Helicone

> 透過 Helicone Gateway 觀察 TokenLab 請求

## 概覽

Helicone 可以透過將 TokenLab 的 OpenAI 相容請求路由至 Helicone Gateway 並指向 TokenLab，來代理並觀察這些請求。

<Note>
  **類型**: 可觀測性閘道 (Observability Gateway)

  **主要路徑**: 透過 Helicone Gateway 進行 OpenAI 相容的聊天補全 (chat completions)

  **支援信心**: 支援的可觀測性路徑
</Note>

## 環境變數

```bash theme={null}
export HELICONE_API_KEY="your-helicone-key"
export TOKENLAB_API_KEY="sk-your-tokenlab-key"
```

## cURL

```bash theme={null}
curl --request POST \
  --url https://gateway.helicone.ai/v1/chat/completions \
  --header "Authorization: Bearer $TOKENLAB_API_KEY" \
  --header "Helicone-Auth: Bearer $HELICONE_API_KEY" \
  --header "Helicone-Target-Url: https://api.tokenlab.sh" \
  --header "Helicone-Target-Provider: TokenLab" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "gpt-5.4-mini",
    "messages": [
      {
        "role": "user",
        "content": "Say hello from TokenLab through Helicone."
      }
    ]
  }'
```

## OpenAI SDK

```typescript theme={null}
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.TOKENLAB_API_KEY,
  baseURL: "https://gateway.helicone.ai/v1",
  defaultHeaders: {
    "Helicone-Auth": `Bearer ${process.env.HELICONE_API_KEY}`,
    "Helicone-Target-Url": "https://api.tokenlab.sh",
    "Helicone-Target-Provider": "TokenLab",
  },
});

const response = await client.chat.completions.create({
  model: "claude-sonnet-5",
  messages: [{ role: "user", content: "Write one concise product summary." }],
});
```

## 範圍

此設定適用於 OpenAI 相容的 `/v1` 流量。TokenLab 同時也公開了原生 Responses、Anthropic Messages 以及 Gemini 相容的端點；除非您的可觀測性代理明確支援這些格式，否則請針對這些原生格式使用直接的 TokenLab 路由。