Official

anthropic / claude-3.7-sonnet

The most intelligent Claude model and the first hybrid reasoning model on the market (claude-3-7-sonnet-20250219)

Warm

Public
885.6K runs
Priced per token
Commercial use
License

Run with an API

Playground API Examples README

Input

Run this model in Node.js with one line of code:

npx create-replicate --model=anthropic/claude-3.7-sonnet

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run anthropic/claude-3.7-sonnet using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const input = {
  prompt: "Give me a recipe for pancakes that could feed all of California.",
  max_tokens: 8192,
  system_prompt: "",
  max_image_resolution: 0.5
};

for await (const event of replicate.stream("anthropic/claude-3.7-sonnet", { input })) {
  process.stdout.write(event.toString());
};

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run anthropic/claude-3.7-sonnet using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

# The anthropic/claude-3.7-sonnet model can stream output as it's running.
for event in replicate.stream(
    "anthropic/claude-3.7-sonnet",
    input={
        "prompt": "Give me a recipe for pancakes that could feed all of California.",
        "max_tokens": 8192,
        "system_prompt": "",
        "max_image_resolution": 0.5
    },
):
    print(str(event), end="")

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run anthropic/claude-3.7-sonnet using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "input": {
      "prompt": "Give me a recipe for pancakes that could feed all of California.",
      "max_tokens": 8192,
      "system_prompt": "",
      "max_image_resolution": 0.5
    }
  }' \
  https://api.replicate.com/v1/models/anthropic/claude-3.7-sonnet/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

# Colossal California Pancake Recipe ## Ingredients - 800 million cups all-purpose flour (roughly 100 million pounds) - 400 million cups milk (about 3.2 billion pints) - 200 million eggs (approximately 16.7 million dozen) - 40 million cups sugar - 20 million tablespoons baking powder - 10 million teaspoons salt - 10 million cups butter, melted ## Instructions 1. Locate approximately 10,000 industrial-sized mixing facilities across the state. 2. Coordinate 100,000 volunteer mixing teams. 3. Combine dry ingredients in massive containers using cement mixer-style equipment. 4. In separate containers, whisk together the milk and eggs (perhaps using fire hose systems for circulation). 5. Gradually incorporate wet ingredients into dry ingredients. 6. Cook on 500,000 simultaneously operating extra-large griddles. 7. Serve with approximately 5 million gallons of maple syrup. ## Notes - Cooking time: Approximately 3 weeks with 24/7 operation - Would require borrowing industrial equipment from multiple nations - May cause maple syrup shortages worldwide - Environmental impact statement pending Would you like me to calculate the exact nutritional information for 39 million servings? 😊

{
  "completed_at": "2025-02-25T15:49:55.109073Z",
  "created_at": "2025-02-25T15:49:49.292000Z",
  "data_removed": false,
  "error": null,
  "id": "9f04v4ypdhrme0cn7q6bc4t0sw",
  "input": {
    "prompt": "Give me a recipe for pancakes that could feed all of California.",
    "max_tokens": 8192,
    "system_prompt": "",
    "max_image_resolution": 0.5
  },
  "logs": "Running prediction...",
  "metrics": {
    "input_token_count": 21,
    "tokens_per_second": 53.60948617837659,
    "output_token_count": 311,
    "predict_time": 5.802968394,
    "total_time": 5.817073,
    "time_to_first_token": 0.015861292
  },
  "output": [
    "#",
    " Colossal California",
    " Pancake",
    " Recipe\n\n## Ingredients",
    "\n- 800",
    " million",
    " cups",
    " all",
    "-purpose flour (",
    "roughly",
    " 100",
    " million",
    " pounds",
    ")\n- 400",
    " million cups",
    " milk",
    " (about",
    " 3",
    ".",
    "2",
    " billion p",
    "ints)\n- ",
    "200 million eggs (",
    "approximately 16.",
    "7 million",
    " dozen",
    ")\n- 40",
    " million cups sugar",
    "\n- 20 million",
    " tabl",
    "espoons baking powder",
    "\n- 10 ",
    "million teaspoons salt",
    "\n- 10",
    " million cups butter,",
    " melted",
    "\n\n## Instructions\n1.",
    " Locate",
    " approximately",
    " 10,",
    "000 industrial",
    "-sized mixing facilities",
    " across",
    " the",
    " state.",
    "\n2. Coordinate",
    " ",
    "100,000 volunteer",
    " mixing",
    " teams.\n3.",
    " Combine",
    " dry",
    " ingredients in massive",
    " containers",
    " using",
    " cement",
    " mixer",
    "-",
    "style",
    " equipment.\n4.",
    " In",
    " separate containers, wh",
    "isk together the",
    " milk",
    " an",
    "d eggs (",
    "perhaps",
    " using fire",
    " hose",
    " systems",
    " for",
    " circulation",
    ").\n5.",
    " Gradually incorporate wet",
    " ingredients into dry ingredients.",
    "\n6. Cook",
    " on",
    " ",
    "500",
    ",000 simultaneously",
    " operating extra",
    "-large griddles.",
    "\n7. Serve",
    " with",
    " approximately",
    " 5",
    " million gallons of",
    " maple syrup.",
    "\n\n## Notes",
    "\n- Cooking",
    " time:",
    " Approximately 3 weeks",
    " with",
    " ",
    "24/7 operation",
    "\n- Woul",
    "d require",
    " borrow",
    "ing industrial",
    " equipment",
    " from multiple",
    " nations",
    "\n- May",
    " cause",
    " maple",
    " syrup shortages worldwide",
    "\n- Environmental",
    " impact statement",
    " pending\n\nWoul",
    "d you like me to calculate",
    " the exact nutrit",
    "ional information",
    " for ",
    "39",
    " million serv",
    "ings?",
    " ",
    "😊"
  ],
  "started_at": "2025-02-25T15:49:49.306105Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream-b.svc.ric1.c.replicate.net/v1/streams/hydokvsqdvn3yb3pq7sndxhmwx32kz64yrscbhojdinf5t2fqn7a",
    "get": "https://api.replicate.com/v1/predictions/9f04v4ypdhrme0cn7q6bc4t0sw",
    "cancel": "https://api.replicate.com/v1/predictions/9f04v4ypdhrme0cn7q6bc4t0sw/cancel"
  },
  "version": "hidden"
}

Generated in

5.8 seconds

Input tokens

Output tokens

310

Tokens per second

53.61 tokens / second

Time to first token

16 milliseconds

Tweak it Report

Pricing

This language model is priced by how many input tokens are sent as inputs and how many output tokens are generated.

Type	Per unit	Per $1
Input	$3.00 / 1M tokens or 330K tokens / $1
Output	$15.00 / 1M tokens or 66K tokens / $1

For example, for $10 you can run around 952 predictions where the input is a sentence or two (15 tokens) and the output is a few paragraphs (700 tokens).

Check out our docs for more information about how per-token pricing works on Replicate.

Readme

Claude 3.7 Sonnet

Claude 3.7 Sonnet is Anthropic’s latest frontier large language model. Showing particularly strong improvements in coding and front-end web development.

This Replicate model is built on claude-3-7-sonnet-20250219", hosted on Anthropic and Vertex APIs.

Use cases

Claude 3.7 Sonnet is optimized for the following use cases:

Agentic coding - Claude 3.7 Sonnet is state-of-the-art for agentic coding, and can complete tasks across the entire software development lifecycle—from initial planning to bug fixes, maintenance to large refactors. It offers strong performance in both planning and solving for complex coding tasks, making Claude 3.7 Sonnet an ideal choice to power end-to-end software development processes.
Customer-facing agents - Claude 3.7 Sonnet offers superior instruction following, tool selection, error correction, and advanced reasoning for customer-facing agents and complex AI workflows.
Computer use - Claude 3.7 Sonnet is our most accurate model for computer use, enabling developers to direct Claude to use computers the way people do.
Content generation and analysis - Claude 3.7 Sonnet excels at writing and is able to understand nuance and tone in content to generate more compelling content and analyze content on a deeper level.
Visual data extraction - With Claude 3.7 Sonnet’s robust vision skills, it is the right choice for teams that want to extract raw data from visuals like charts or graphs as part of their AI workflow.

Getting started

To use Claude 3.7 Sonnet via Replicate:

import replicate

output = replicate.run(
    "anthropic/claude-3.7-sonnet",
    input={
        "prompt": "Your prompt here"
    }
)

Privacy policy & license

Data from this model is sent to Anthropic and Google Cloud Vertex AI.

Usage of this model is subject to Anthropic’s terms of service. Please refer to their website for full terms and conditions.