Amazon Bedrock 在首尔和新加坡推出 Anthropic 模型区域内推理

英文标题：Introducing Anthropic models on Amazon Bedrock for in-region inference in Seoul and Singapore

Amazon Bedrock 现已在首尔区域提供 Claude Opus 5 和 Claude Sonnet 5 的区域内推理，并在新加坡区域提供 Claude Sonnet 5 的区域内推理。推理请求、输入提示和输出结果均在调用的单一 AWS Region 内处理，适合金融服务、医疗和公共部门等具有严格数据驻留要求的场景。开发者可通过 `bedrock-runtime` 端点，使用 Anthropic Messages API、Amazon Bedrock InvokeModel API 或 Converse API 调用这些模型。区域内推理不经过跨区域路由层，但吞吐量受限于当地区域容量，并按该区域的按需价格、服务配额和监控范围执行。

1.3 MB 101.1 KB 124.1 KB 345.0 KB 141.2 KB 346.5 KB 92.5 KB 15.1 KB 21.1 KB 15.8 KB 15.2 KB 57.6 KB

import boto3
import json

# Create a Bedrock Runtime client with in-region inference in Singapore
bedrock_runtime = boto3.client(
    service_name="bedrock-runtime",
    region_name="ap-southeast-1")

# Invoke Claude Sonnet 5
response = bedrock_runtime.invoke_model(
    modelId="anthropic.claude-sonnet-5",
    contentType="application/json",
    accept="application/json",
    body=json.dumps({
        "anthropic_version": "bedrock-2023-05-31",
        "max_tokens": 4096,
        "messages": [
            {
                "role": "user",
                "content": " Can you explain the features of Amazon Bedrock? "
            }
        ]
    })
)

result = json.loads(response["body"].read())
print(result["content"][0]["text"])

import boto3

# Create a Bedrock Runtime client with in-region inference in Seoul
bedrock_runtime = boto3.client(
    service_name="bedrock-runtime",
    region_name="ap-northeast-2"
)

# Invoke Claude Opus 5
response = bedrock_runtime.converse(
    modelId="anthropic.claude-opus-5",
    messages=[
        { "role": "user",
          "content": [
              { "text": " Can you explain the features of Amazon Bedrock?"
              }
          ]
        }
    ],
    inferenceConfig={
        "maxTokens": 4096
    }
)

if 'output' in response:
    blocks = response['output']['message']['content']
    print('\\n'.join(b.get('text', '') for b in blocks if 'text' in b))

from anthropic import Anthropic
from aws_bedrock_token_generator import provide_token

token = provide_token(region="ap-northeast-2")

client = Anthropic(
    base_url="https://bedrock-runtime.ap-northeast-2.amazonaws.com/anthropic",
    api_key=token,
)

response = client.messages.create(
    model="anthropic.claude-sonnet-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Can you explain the features of Amazon Bedrock?"}],
)

print(response)

import boto3
import json

# Create a Bedrock Runtime client with in-region inference in Singapore
bedrock_runtime = boto3.client(
    service_name="bedrock-runtime",
    region_name="ap-southeast-1")

# Invoke Claude Sonnet 5
response = bedrock_runtime.invoke_model(
    modelId="anthropic.claude-sonnet-5",
    contentType="application/json",
    accept="application/json",
    body=json.dumps({
        "anthropic_version": "bedrock-2023-05-31",
        "max_tokens": 4096,
        "messages": [
            {
                "role": "user",
                "content": " Can you explain the features of Amazon Bedrock? "
            }
        ]
    })
)

result = json.loads(response["body"].read())
print(result["content"][0]["text"])

import boto3

# Create a Bedrock Runtime client with in-region inference in Seoul
bedrock_runtime = boto3.client(
    service_name="bedrock-runtime",
    region_name="ap-northeast-2"
)

# Invoke Claude Opus 5
response = bedrock_runtime.converse(
    modelId="anthropic.claude-opus-5",
    messages=[
        { "role": "user",
          "content": [
              { "text": " Can you explain the features of Amazon Bedrock?"
              }
          ]
        }
    ],
    inferenceConfig={
        "maxTokens": 4096
    }
)

if 'output' in response:
    blocks = response['output']['message']['content']
    print('\n'.join(b.get('text', '') for b in blocks if 'text' in b))

from anthropic import Anthropic
from aws_bedrock_token_generator import provide_token

token = provide_token(region="ap-northeast-2")

client = Anthropic(
    base_url="https://bedrock-runtime.ap-northeast-2.amazonaws.com/anthropic",
    api_key=token,
)

response = client.messages.create(
    model="anthropic.claude-sonnet-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Can you explain the features of Amazon Bedrock?"}],
)

print(response)