Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

.Net: Support Function calling for Amazon Bedrock (Using Converse APIs) #11448

Open
RogerBarreto opened this issue Apr 8, 2025 · 1 comment
Labels
ai connector Anything related to AI connectors function_calling .NET Issue or Pull requests regarding .NET code triage

Comments

@RogerBarreto
Copy link
Member

Discussed in #11436

Originally posted by breggles April 8, 2025
Hi,

I was trying out Claude 3.7 Sonnet on Amazon Bedrock and function calling wasn't working, when it was working fine with Azure OpenAI. I then found this table that says it's not even on the list of things you're aware of. Do you know when you will support function calling for Amazon Bedrock? Or Anthropic, or Gemini for that matter?

TIA,

Tilman

Our current implementation of Bedrock doesn't have tool support for the specific Amazon's Bedrock Tool Converse pattern below:

tool-use-inference-call

This ticket covers this feature request.

@RogerBarreto RogerBarreto added .NET Issue or Pull requests regarding .NET code ai connector Anything related to AI connectors function_calling triage labels Apr 8, 2025
@breggles
Copy link

breggles commented Apr 8, 2025

Hey, thanks for turning this into an issue. And thanks for a great product! I just wanted to add that these days function/tool calling is essential for pretty much any AI app, and without it pictures like these:

Image

and texts like these:

Semantic Kernel was designed to be future proof, easily connecting your code to the latest AI models evolving with the technology as it advances. When new models are released, you’ll simply swap them out without needing to rewrite your entire codebase.

are a bit of a lie, because you can't just swap out your model for a new one. I appreciate it's a lot of work and you can't support all models/hosts, but without the ability to swap models easily, you're losing a lot of the value proposition of Semantic Kernel.

(Picture and text from here: https://learn.microsoft.com/en-us/semantic-kernel/overview/)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ai connector Anything related to AI connectors function_calling .NET Issue or Pull requests regarding .NET code triage
Projects
None yet
Development

No branches or pull requests

2 participants