Note

As of the time of creating the Repo, Claude Anthropic and OpenAI API's do not support PDF upload, therefore we need to convert and split the PDF that we want to have Analyzed into multiple images, and send them to the API's.
This is not supported in Spring AI as well.
In the future all of the things mentioned above will be probably changed and PDF uploads will be directly supported.

How to use this project?

In order to use this project, you need to register for OpenAI and Claude Anthropic and add credit to your accounts.

1. OpenAI link - https://platform.openai.com/settings/organization/billing/overview
2. Claude Anthropic link - https://console.anthropic.com/settings/billing
3. Generate API keys for both of them through their consoles

After that you need to set the following environment variables (the API keys) in order to run the project successfully:

1. OPENAI_API_KEY
2. ANTHROPIC_API_KEY

Start the project and you can use Postman/cURL to test the two endpoints:

1. Test OpenAI endpoint

curl --location 'http://localhost:8080/api/v1/upload/openai/pdf' \
--form 'file=@"/E:/Projects/spring-ai/documents/example-rate-confirmation.pdf"'

2. Test Claude Anthropic endpoint

curl --location 'http://localhost:8080/api/v1/upload/anthropic/pdf' \
--form 'file=@"/E:/Projects/spring-ai/documents/example-rate-confirmation.pdf"'

Materials to use

Several good usages of the Spring AI API

https://github.com/danvega/spring-into-ai
https://spring.io/blog/2024/05/30/spring-ai-1-0-0-m1-released
https://docs.spring.io/spring-ai/reference/api/chat/openai-chat.html

How to convert pdf to images

https://stackoverflow.com/questions/23326562/convert-pdf-files-to-images-with-pdfbox
LibrePDF/OpenPDF#346 - not working, just posting the link to track, using PDFBox instead

Issues

https://stackoverflow.com/questions/78121525/spring-ai-opentezi teai-error-extracting-response-of-type-openaiapiembeddinglist
https://stackoverflow.com/questions/15414259/java-bufferedimage-to-byte-array-and-back

How to use Anthropic

https://www.danvega.dev/blog/claude-sonnet-spring-ai
https://docs.anthropic.com/en/api/getting-started
https://docs.anthropic.com/en/api/messages-examples#vision
https://github.com/danvega/hello-claude

How to use Multiple LLMs with Spring Boot

https://www.danvega.dev/blog/spring-ai-multiple-llms