LangChain

PulseNeko2026/5/20小于 1 分钟

LangChain 接 Bridge 的核心就一行：把 LLM 实例的 base_url 改掉。

langchain-openai

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="gpt-5.1",
    base_url="https://bridge.pulseneko.com/v1",
    api_key="sk-你的-Key",
)

print(llm.invoke("Hello").content)

环境变量方式：

export OPENAI_API_KEY=sk-你的-Key
export OPENAI_BASE_URL=https://bridge.pulseneko.com/v1

然后代码里不用再传 api_key / base_url。

langchain-anthropic

from langchain_anthropic import ChatAnthropic

llm = ChatAnthropic(
    model="claude-opus-4-7",
    base_url="https://bridge.pulseneko.com",
    api_key="sk-你的-Key",
    max_tokens=1024,
)

print(llm.invoke("Hello").content)

LangGraph

LangGraph 用上面任意一个 LLM 实例即可。Agent 流程内部多次调用都会用同一个 Bridge Key。

Embeddings

from langchain_openai import OpenAIEmbeddings

emb = OpenAIEmbeddings(
    model="text-embedding-3-large",
    base_url="https://bridge.pulseneko.com/v1",
    api_key="sk-你的-Key",
)

LlamaIndex

from llama_index.llms.openai import OpenAI

llm = OpenAI(
    model="gpt-5.1",
    api_base="https://bridge.pulseneko.com/v1",
    api_key="sk-你的-Key",
)

LlamaIndex 用 api_base（不是 base_url），注意区分。

Vercel AI SDK

import { createOpenAI } from "@ai-sdk/openai";

const bridge = createOpenAI({
  baseURL: "https://bridge.pulseneko.com/v1",
  apiKey: process.env.PULSENEKO_KEY,
});

const result = await generateText({
  model: bridge("gpt-5.1"),
  prompt: "Hello",
});