Anthropic (Claude)

Overview

Anthropic is an AI safety company building the Claude family of models, known for strong reasoning, long context, vision/PDF understanding, and a rich set of agentic tools.

Official Website: https://www.anthropic.com API Documentation: https://docs.anthropic.com

Key features

• Extended thinking — a configurable thinking budget for deeper reasoning
• Vision & PDF — analyze images and documents inline
• Tool use — function calling for agentic workflows
• Prompt caching — cache large, reused prompt prefixes to cut cost and latency
• Batch — high-volume asynchronous processing at a discount
• Server tools — built-in web_search, web_fetch, code_execution, computer_use, and the MCP connector

Using Anthropic with Yunxin

Address Claude models by prefixing the model id with the anthropic slug — anthropic/<model> (for example anthropic/claude-...). The recommended way to call Claude is the Anthropic-native Messages API at POST /v1/messages, which preserves Claude-specific features. The anthropic-version header is accepted (e.g. anthropic-version: 2023-06-01).

You always use the same Yunxin endpoint with your sk- API key. With the Anthropic SDK, set base_url to https://api.yuhuanstudio.com (the SDK appends /v1/messages itself).

Usage example

import os
from anthropic import Anthropic

client = Anthropic(
    api_key=os.environ["YUNXIN_API_KEY"],
    base_url="https://api.yuhuanstudio.com",
)

message = client.messages.create(
    model="anthropic/claude-model-id",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Hello, Claude!"},
    ],
)

print(message.content[0].text)

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

const client = new Anthropic({
  apiKey: process.env.YUNXIN_API_KEY,
  baseURL: "https://api.yuhuanstudio.com",
});

const message = await client.messages.create({
  model: "anthropic/claude-model-id",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Hello, Claude!" }],
});

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

curl https://api.yuhuanstudio.com/v1/messages \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $YUNXIN_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "anthropic/claude-model-id",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Hello, Claude!"}
    ]
  }'

Extended thinking

Enable a thinking budget for complex problems:

message = client.messages.create(
    model="anthropic/claude-model-id",
    max_tokens=2048,
    thinking={"type": "enabled", "budget_tokens": 10000},
    messages=[{"role": "user", "content": "Solve this step by step."}],
)

See Reasoning for how thinking is surfaced across formats.

Available models

List the models currently available for this provider:

curl "https://api.yuhuanstudio.com/v1/models?provider=anthropic" \
  -H "Authorization: Bearer $YUNXIN_API_KEY"

Official resources

• Anthropic Website
• API Documentation
• Console
• Pricing