Qwen 3.8 — AEON Ultimate Uncensored - a AEON-7 Collection

AEON-7 's Collections

Gemma 4 — DECKARD HERETIC, Multimodal & Speculators

Qwen 3.8 — AEON Ultimate Uncensored

Qwen 3.5 — DFlash Drafters

Nemotron 3 Nano Omni — AEON Ultimate Uncensored

Step-3.7-Flash AEON-Ultimate (Abliterated)

Qwen 3.8 — AEON Ultimate Uncensored

updated 6 days ago

Qwen3.8-27B AEON Ultimate Uncensored — BF16 master + NVFP4-MIXED deploy (vision + MTP) + MXFP4/MXFP6 ROCm early release (gated).

AEON-7/Qwen3.8-27B-AEON-ULTIMATE-UNCENSORED-BF16

Text Generation • 28B • Updated 6 days ago • 13.8k • • 223

Note BF16 master — full-precision uncensored teacher.
AEON-7/Qwen3.8-27B-AEON-ULTIMATE-UNCENSORED-NVFP4-MIXED

Text Generation • 20B • Updated 6 days ago • 9.3k • 45

Note NVFP4-MIXED deploy — vision + native MTP, Spark/5090/PRO6000.
AEON-7/Qwen3.8-27B-AEON-ULTIMATE-UNCENSORED-MXFP4-MXFP6-ROCm

Text Generation • 18B • Updated 6 days ago • 10 • 1

Note MXFP4/MXFP6 ROCm — Early Release · Gated Access. Radeon AI PRO R9700 (emulated MX) · Instinct MI350/MI355 (native MXFP4).
AEON-7/AEON-DFlash2-Qwen3.8-27B

Text Generation • 1B • Updated 6 days ago • 32 • 2

Note AEON DFlash2 v1.0 (lk4) drafter: Early Release · Gated Access. +21% over stock z-lab DFlash2 at T=0.6 on DGX Spark (measured with lk3). NVFP4 + BF16.