wepiqx author hub
Ornith 1.5 9B / 1.0 9B — ASHQ1 Quantization MTP Aware ASHQ1 quantization of Ornith with explicit MTP Multi Token Prediction handling — the whole MTP head blk.32. + nextn. is kept at Q8 0 and excluded from the classifier budget, while the priority queue optimises the main transfo…
Models
15
Downloads
9,845
Run models locally with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.