Feature Request: OpenResponses API

[njbrake]

Prerequisites

• I am running the latest code. Mention the version if possible as well.
• I carefully followed the README.md.
• I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
• I reviewed the Discussions, and have a new and useful enhancement to share.

Feature Description

Ball is rolling to get llama.cpp and mlx-lm to support OpenResponses protocol.

ml-explore/mlx-lm#815

ggml-org/llama.cpp#19138

Is there anything else that needs to be done for llamafile to support OpenResponses protocol once llama.cpp is updated?

Motivation

The ChatCompletion API was not designed for agentic use.

Possible Implementation

No response

[aittalam]

Hi and thanks for the great idea! I don't think we should do anything particular once llama.cpp is updated (except syncing with upstream), as we are now relying on their server mode. The sync itself is now a bit easier than before, so it should not be too painful (see my update here).

I see there's an open PR for it so let's keep an eye on that!