Feature Request: deep/ recurrent processing like "thinking", but script based.

[David-AU-github]

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

Curious if this model could be supported/GGUFed with very different processing methods:

https://huggingface.co/tomg-group-umd/huginn-0125

(F32, Transformers, 3.71B => But deep/ recurrent processing like "thinking")

Paper:
https://www.arxiv.org/abs/2502.05171

Motivation

Unique method of "thinking" / "processing" prompts.

Possible Implementation

Based on inspection, this would require loop(s) of prompt processing, much like "thinking"/"reasoning" models, but contained within the "server" before it "spits out" the response.

[github-actions]

This issue was closed because it has been inactive for 14 days since being marked as stale.