admin管理员组

文章数量:1402943

  1. I will have documents like PDF, JPEG, PNG stored on s3.
  2. I want to post them to Anthropic claude for OCR processing.
  3. Instead of using boto3 to download and convert them to base64 and send bytes to to bedrock invoke method. Is there a way where I can directly specify the s3 URL of document and bedrock takes care of it? For example
bedrock_agent_runtime.invoke(modelID,s3Location="s3://bucket_name/w2.pdf")

I didn't find any API for this.

  1. I will have documents like PDF, JPEG, PNG stored on s3.
  2. I want to post them to Anthropic claude for OCR processing.
  3. Instead of using boto3 to download and convert them to base64 and send bytes to to bedrock invoke method. Is there a way where I can directly specify the s3 URL of document and bedrock takes care of it? For example
bedrock_agent_runtime.invoke(modelID,s3Location="s3://bucket_name/w2.pdf")

I didn't find any API for this.

  • amazon-web-services
  • amazon-s3
  • amazon-bedrock
Share Improve this question edited Mar 23 at 17:13 desertnaut 60.5k32 gold badges155 silver badges181 bronze badges asked Mar 21 at 1:23 RajRaj 4177 silver badges25 bronze badges
Add a comment  | 

1 Answer 1

Reset to default 0

Apologies for the previous link. It was incorrect.

Please take a look at these examples. They use Bedrock to send through the contents of an S3 object through to the model.
I do not see the option to send the ARN/URL of the S3 Object itself as a method input parameter.

本文标签： amazon web servicespost s3 document URL to bedrock llmStack Overflow

版权声明：本文标题：amazon web services - post s3 document URL to bedrock llm - Stack Overflow 内容由网友自发贡献，该文观点仅代表作者本人， 转载请联系作者并注明出处：http://www.betaflare.com/web/1744376297a2603276.html， 本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌抄袭侵权/违法违规的内容，一经查实，本站将立刻删除。