license: cc-by-4.0 language:
- en tags:
- mixture of experts
- moe
- 8x3B
- Llama 3.2 MOE
- 128k context
- creative
- creative writing
- fiction writing
- plot generation
- sub-plot generation
- fiction writing
- story generation
- scene continue
- storytelling
- fiction story
- science fiction
- romance
- all genres
- story
- writing
- vivid prosing
- vivid writing
- fiction
- roleplaying
- bfloat16
- swearing
- rp
- horror
- mergekit
- llama
- llama-3
- llama-3.2
- heretic
- uncensored
- decensored
- abliterated
- finetune
- gguf
- llama.cpp
- rocmfpx
- rocmfp4 pipeline_tag: text-generation library_name: llama.cpp
DavidAU/Llama3.2-24B-A3B-II-Dark-Champion-INSTRUCT-Heretic-Abliterated-Uncensored (Q4_0_ROCmFP4_FAST)
- Original Unquantized Model: DavidAU/Llama3.2-24B-A3B-II-Dark-Champion-INSTRUCT-Heretic-Abliterated-Uncensored
- Quantization Format:
Q4_0_ROCmFP4_FAST - Required Backend: charlie12345/ROCmFPX
Overview
This repository provides an optimized GGUF file compiled using custom ROCmFPX memory layouts designed for high-performance hardware acceleration on AMD ROCm architectures.
Requirements & Compatibility
Do not attempt to execute this file using standard upstream llama.cpp, KoboldCPP, or LM Studio unless the underlying binaries have been explicitly compiled against or patched with the charlie12345/ROCmFPX repository extensions.