Anthropic SDK

PulseNeko5/20/26About 1 min

The native /v1/messages protocol gives Claude family models full fidelity — prompt cache, tool use, document blocks.

Python

import anthropic

client = anthropic.Anthropic(
    base_url="https://bridge.pulseneko.com",
    api_key="sk-your-key",
)

msg = client.messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)
print(msg.content[0].text)

Don't append /v1

The Anthropic SDK appends /v1/messages internally. Set base URL to https://bridge.pulseneko.com. Adding /v1 yourself yields /v1/v1/messages → 404.

Auth

The Anthropic SDK reads ANTHROPIC_API_KEY by default and sends it as x-api-key. Bridge accepts both x-api-key and Authorization: Bearer.

Streaming

with client.messages.stream(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[{"role": "user", "content": "..."}],
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

Prompt cache

msg = client.messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    system=[
        {
            "type": "text",
            "text": "<long system prompt>",
            "cache_control": {"type": "ephemeral"},
        }
    ],
    messages=[{"role": "user", "content": "..."}],
)

Cache hits are billed at the Cache column on the Model Catalog — usually much lower than Input.

Tool use

msg = client.messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    tools=[{
        "name": "get_weather",
        "description": "Look up weather",
        "input_schema": {
            "type": "object",
            "properties": {"city": {"type": "string"}},
            "required": ["city"],
        },
    }],
    messages=[{"role": "user", "content": "What's the weather in Shanghai?"}],
)

Node.js

import Anthropic from "@anthropic-ai/sdk";

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

const msg = await client.messages.create({
  model: "claude-opus-4-7",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Hello" }],
});

console.log(msg.content[0].text);

Recommended models

Model	Best for
`claude-haiku-4-5`	High QPS, low latency, cheap
`claude-sonnet-4-6`	Daily driver
`claude-opus-4-7`	High-quality reasoning, long context
`claude-opus-4-7-thinking`	4.7 + extended thinking