By using AWS re:Post, you agree to the AWS re:Post Terms of Use
AWS re:Postre:Post

Bedrock: Claude Model Selection (Context Window)

0

How can I determine which modelID is used when using on-demand Amazon Bedrock? For example, Claude 2.1 has 18k and 200k context window model IDs. Can I just provide the context window and count on the system to choose the cheapest model?

Thanks

Topics
Machine Learning & AIGenerative AI on AWS
Tags
Machine Learning & AIAmazon Bedrock
Language
English
rePost-User-5941005
asked 7 months ago1069 views
1 Answer
1
Accepted Answer

Amazon Bedrock model IDs shows the model IDs. Below is a copy of the Anthropic models as of this answer. There is only one Claude 2.1 available (anthropic.claude-v2:1). From https://aws.amazon.com/bedrock/claude/ we see this is the one "with a 200K token context window".

ProviderModel nameVersionModel ID
AnthropicClaude2.0anthropic.claude-v2
AnthropicClaude2.1anthropic.claude-v2:1
AnthropicClaude 3 Sonnet1.0anthropic.claude-3-sonnet-20240229-v1:0
AnthropicClaude 3 Haiku1.0anthropic.claude-3-haiku-20240307-v1:0
AnthropicClaude Instant1.xanthropic.claude-instant-v1

The Anthropic Claude Text Completions API provides inference parameters, including max_tokens_to_sample (maximum number of tokens to generate before stopping). There is no parameter for input tokens / context window, since you control that via the prompt parameter.

profile pictureAWS
hakanson
answered 7 months ago
profile picture
EXPERT
Oleksii Bebych
reviewed 5 months ago
profile picture
EXPERT
Giovanni Lauria
reviewed 7 months ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions

Relevant content