Claude Opus 4.7 API

Model IDs by platform

Basic usage — Python

import anthropic

client = anthropic.Anthropic()

message = client.messages.create(
    model="claude-opus-4-7-20260416",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": "Review this PR diff and identify any bugs."
        }
    ]
)

print(message.content[0].text)

Basic usage — TypeScript

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

const client = new Anthropic();

const message = await client.messages.create({
  model: "claude-opus-4-7-20260416",
  max_tokens: 1024,
  messages: [
    {
      role: "user",
      content: "Review this PR diff and identify any bugs."
    }
  ]
});

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

Effort levels

Opus 4.7 adds a new xhigh effort level between high and max.

# Using xhigh effort for hard coding tasks
message = client.messages.create(
    model="claude-opus-4-7-20260416",
    max_tokens=16000,
    thinking={
        "type": "enabled",
        "effort": "xhigh"   # new in Opus 4.7
    },
    messages=[{"role": "user", "content": prompt}]
)

Prompt caching — 90% off input costs

Cache your system prompt, large documents, or tool definitions. Cached tokens cost $0.50/1M instead of $5/1M.

message = client.messages.create(
    model="claude-opus-4-7-20260416",
    max_tokens=1024,
    system=[
        {
            "type": "text",
            "text": "You are a senior software engineer...",
            "cache_control": {"type": "ephemeral"}  # cache this block
        }
    ],
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "text",
                    "text": large_codebase_context,   # also cache this
                    "cache_control": {"type": "ephemeral"}
                },
                {
                    "type": "text",
                    "text": "Now review the following diff..."
                }
            ]
        }
    ]
)

Vision — 98.5% accuracy, 2,576px images

Opus 4.7 handles images up to 2,576px on the long edge. Pass them as base64 or via URL.

import base64

with open("chart.png", "rb") as f:
    image_data = base64.b64encode(f.read()).decode("utf-8")

message = client.messages.create(
    model="claude-opus-4-7-20260416",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "image",
                    "source": {
                        "type": "base64",
                        "media_type": "image/png",
                        "data": image_data,
                    },
                },
                {
                    "type": "text",
                    "text": "Analyse this financial chart and summarise the trend."
                }
            ],
        }
    ],
)

Tokenizer migration checklist

Before moving production traffic from Opus 4.6 to 4.7.

Related