Qwen3.8-27B-ULT3-QNext-Stage1.1: Model Card

This model is part of this project:

https://huggingface.co/DavidAU/Qwen3.8-27B-UltimateDetails2-stage1__The-Harley-Pelican

This is a work in progress, if you want to be notified the FINAL release, go to ABOVE REPO to request access.

• Stage 1 built, tested.
• Stage 1.1 and Stage 1.5 => Built, in testing.

Benches: [see full project notes/benches at above repo]

          arc/c arc/e boolq hswag obkqa piqa  wino

[NEW BRANCH CORE START]

Ultimate 3 [Qwen Next Branch; ground up build, components first, then combine.]
Qwen3.8-27B-ULT3-QNext-Stage1 => Qnext Reasoning Only Training Mild.
mxfp8     0.617,0.809,0.906,...

Stage 1.1 :
Qwen3.8-27B-ULT3-QNext-Stage1.1
mxfp8    0.622,0.801,0.901,...

Stage 1.5
- pending


[QWENS]
[base, non heretic, untuned]

Qwen3.8-27B-Instruct: 
mxfp8     0.591,0.782,0.896,0.746,0.448,0.801,0.711

Qwen3.6-27B-Instruct: 
mxfp8     0.647,0.803,0.910,0.773,0.450,0.806,0.742

Qwen3.5-27B-Instruct: 
mxfp8     0.557,0.711,0.868,0.533,0.452,0.706,0.695

NOTES: - Models are tested in "Instruct" mode because this generally works better with the testing harness. - Testing via "thinking" mode also shows the metrics (and changes) but not the true extent. - In actual fact when the model IS in thinking mode, it will exceed INSTRUCT benchmark scores in most cases. - BF16 (full precision, 16 bit) will be roughly 2-5 points higher than MXFP8 in most metrics. Some metrics may be slightly higher than this.