Product

Introducing the Message Batches API

Oct 8, 2024●2 min read

Illustration of a hand reaching for a batch of requests

We’re introducing a new Message Batches API—a powerful, cost-effective way to process large volumes of queries asynchronously.

Developers can send batches of up to 10,000 queries per batch. Each batch is processed in less than 24 hours and costs 50% less than standard API calls. This makes processing non-time-sensitive tasks more efficient and cost-effective.

The Batches API is available today in public beta with support for Claude 3.5 Sonnet, Claude 3 Opus, and Claude 3 Haiku on the Anthropic API. Customers using Claude in Amazon Bedrock can use batch inference. Support for batch processing for Claude on Google Cloud’s Vertex AI is coming soon.

High throughput at half the cost

Developers often use Claude to process vast amounts of data—from analyzing customer feedback to translating languages—where real-time responses aren't necessary.

Instead of managing complex queuing systems or worrying about rate limits, you can use the Batches API to submit groups of up to 10,000 queries and let Anthropic handle the processing at a 50% discount. Batches will be processed within 24 hours, though often much quicker. Additional benefits include:

Enhanced throughput: Enjoy higher rate limits to process much larger request volumes without impacting your standard API rate limits.
Scalability for big data: Handle large-scale tasks such as dataset analysis, classification of large datasets, or extensive model evaluations without infrastructure concerns.

The Batches API unlocks new possibilities for large-scale data processing that were previously less practical or cost-prohibitive. For example, analyzing entire corporate document repositories—which might involve millions of files—becomes more economically viable by leveraging our batching discount.

Pricing

The Batches API allows you to take advantage of infrastructure cost savings and is offered at a 50% discount for both input and output tokens.

Claude 3.5 Sonnet Our most intelligent model to date 200K context window	Batch Input $1.50 / MTok	Batch Output $7.50 / MTok
Claude 3 Opus Powerful model for complex tasks 200K context window	Batch Input $7.50 / MTok	Batch Output $37.50 / MTok
Claude 3 Haiku Fastest, most cost-effective model 200K context window	Batch Input $0.125 / MTok	Batch Output $0.625 / MTok

Customer Spotlight: Quora

Quora, a user-based question-and-answer platform, leverages Anthropic's Batches API for summarization and highlight extraction to create new end-user features.

"Anthropic's Batches API provides cost savings while also reducing the complexity of running a large number of queries that don't need to be processed in real time," said Andy Edmonds, Product Manager at Quora. "It's very convenient to submit a batch and download the results within 24 hours, instead of having to deal with the complexity of running many parallel live queries to get the same result. This frees up time for our engineers to work on more interesting problems.”

Get started

To start using the Batches API in public beta on the Anthropic API, explore our documentation and pricing page.