OpenAI SDK 接入 GPUShare

通用最广 —— OpenAI SDK 可调任意 30+ 模型 (含 Claude / Gemini / GLM / Grok / ...)

端点

项	值
Base URL	`https://api.dflop.top/v1`
端点	`/chat/completions` (兼容 OpenAI Chat Completions)
鉴权	`Authorization: Bearer sk-gpushare-xxx`
协议	OpenAI Chat Completions (HTTP / SSE 流式)

安装

pip install openai          # Python
npm install openai          # TypeScript

Python

基础调用

from openai import OpenAI

client = OpenAI(
    base_url="https://api.dflop.top/v1",
    api_key="sk-gpushare-xxx",
)

resp = client.chat.completions.create(
    model="gpt-5.4",
    messages=[{"role": "user", "content": "Say hello in one word."}],
)
print(resp.choices[0].message.content)

流式输出

stream = client.chat.completions.create(
    model="gpt-5.4",
    messages=[{"role": "user", "content": "Stream a haiku about latency"}],
    stream=True,
)
for chunk in stream:
    delta = chunk.choices[0].delta.content or ""
    print(delta, end="", flush=True)

跨厂商调用 (重点)

OpenAI SDK 可以调用任意支持 OpenAI Chat 协议的模型 —— 不只是 GPT:

# 调 Claude
resp = client.chat.completions.create(
    model="claude-sonnet-4-5-20250929",
    messages=[{"role": "user", "content": "Hello"}],
)

# 调 Gemini
resp = client.chat.completions.create(
    model="gemini-2.5-pro",
    messages=[{"role": "user", "content": "Hello"}],
)

# 调 GLM
resp = client.chat.completions.create(
    model="glm-5.1",
    messages=[{"role": "user", "content": "Hello"}],
)

图像生成 (sub2api WS V2)

tools=[{"type":"image_generation"}] 必须配合 stream=True:

stream = client.chat.completions.create(
    model="gpt-5.4",
    messages=[{"role": "user", "content": "Draw an orange cat in pixel art"}],
    tools=[{"type": "image_generation"}],
    stream=True,
)

# 图像以 base64 markdown 内联返回: ![](data:image/png;base64,...)
for chunk in stream:
    delta = chunk.choices[0].delta.content or ""
    print(delta, end="", flush=True)

联网搜索

stream = client.chat.completions.create(
    model="gpt-5.4",
    messages=[{"role": "user", "content": "What did Anthropic ship this week?"}],
    tools=[{"type": "web_search"}],
    stream=True,
)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="", flush=True)

Function Tools

跟 OpenAI 行为一致。走 HTTP 通道,不经过 WS V2,stream:false 也支持:

tools = [{
    "type": "function",
    "function": {
        "name": "get_weather",
        "description": "Get the current weather in a city",
        "parameters": {
            "type": "object",
            "properties": {
                "city": {"type": "string"},
            },
            "required": ["city"],
        },
    },
}]

resp = client.chat.completions.create(
    model="claude-sonnet-4-5-20250929",
    messages=[{"role": "user", "content": "Weather in Tokyo?"}],
    tools=tools,
)
print(resp.choices[0].message.tool_calls)

TypeScript

基础调用

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.GPUSHARE_API_KEY,
  baseURL: "https://api.dflop.top/v1",
});

const resp = await client.chat.completions.create({
  model: "gpt-5.4",
  messages: [{ role: "user", content: "Say hello in one word." }],
});

console.log(resp.choices[0].message.content);

流式输出

const stream = await client.chat.completions.create({
  model: "claude-sonnet-4-5-20250929",
  messages: [{ role: "user", content: "Stream a haiku" }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}

跨厂商调用

// 调 Claude
await client.chat.completions.create({
  model: "claude-opus-4-5-20251101",
  messages: [{ role: "user", content: "Hello" }],
});

// 调 Grok
await client.chat.completions.create({
  model: "grok-4-fast-reasoning",
  messages: [{ role: "user", content: "Hello" }],
});

curl

非流式

curl https://api.dflop.top/v1/chat/completions \
  -H "Authorization: Bearer $GPUSHARE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.4",
    "messages": [
      {"role": "user", "content": "Say hello in one word."}
    ]
  }'

流式

curl https://api.dflop.top/v1/chat/completions \
  -H "Authorization: Bearer $GPUSHARE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-5-20250929",
    "stream": true,
    "messages": [
      {"role": "user", "content": "Stream a haiku"}
    ]
  }' \
  --no-buffer

注意事项

stream: false 仅支持纯文本 + function 工具。web_search / image_generation 必须 stream: true
Function 工具 (type: function) 跟 OpenAI 完全一致 —— HTTP 通道,任何模型都支持
错误格式 统一返回 OpenAI 风格: {"error": {"message": "...", "type": "...", "code": "..."}}
配额耗尽 返回 HTTP 412 + quota_exhausted
不支持组合: 暂无 (OpenAI Chat 端点支持所有公开模型)

完整模型列表

模型列表或兼容矩阵