Rogue Creative Series - GGUF / Incl Source too. - a DavidAU Collection

DavidAU 's Collections

Dark Planet Series (see "source" coll. for FP)

Grand Horror 16.5B+ Horror and Fiction Generation

Dark MOEs - Mixture of Experts - Uncensored Creative Models

Dark Champion Collection MOE - Mixture Of Experts

200+ Roleplay, Creative Writing, Uncensored, NSFW models.

OpenAI-GPT 20B, 37B ,120B: Neo, reg, uncensored, ablit.

Dark / Evil / NSFW Reasoning Models (gguf/source)

100 Coder/Programming - MOE, Reasoning, Reg, Imatrix, Fused.

Coders and Programmers: 128 expert / A3B Coders

Coders and Programmers: 1x to 10x MOE - Mixture of Experts

Coders and Programmers: 11B and 12B Fused Models

Coders and Programmers: 4B Models with Brainstorm 20x (6B)

Coders and Programmers: 0.8B, 2.4B, 3.4B

Qwen3 - 30B-A3B (128 experts) and higher

Qwen3 - 16B-A3B - 64 experts + Brainstorm versions

Thinking / Reasoning Models - Reg and MOEs.

MOE - Reasoning - Gated IQ Multi-Tier Models

MOE/Mixture of Experts Models (see also "source" cll)

Qwen 3 / 2.5 Reasoning/Thinking REG + MOEs.

Qwen 3 - Horror / Neo Imatrix / Max Qs / 32-256k ctx

Gemma The Writer Series - GGUF / Source)

Gemma3 - Modified / Augumented Models

Rogue Creative Series - GGUF / Incl Source too.

Brainstorm Adapter Models - Augmented/Expanded Reasoning

David's Software, DOCs and How To for Models.

Higher Precision GGUFs / Imatrix Plus

X-Quants - State of Mind Adjusted Quants

Long Context - 16k,32k,64k,128k,200k,256k,512k,1000k

Instruct Models - Better instruction following.

10B / 10.7B / 11B models (except MOEs)

WS - Qwen3 4B - Fiction on Fire - Merge/pruning series.

WS - Dark Planet Wordstorm Project - Random Prune / Form.

WS - Wordstorm 10 Part Series incl Full Source

Source files for GGUF, EXL2, AWQ, GPTQ, HQQ etc etc

Reasoning Source files for GGUF, EXL2, AWQ, GPTQ

Reasoning Adapters / LORAs -> Any model to reasoning

Experiments in Merging Top Models

Older models - 1b,3b,4b - Upgraded Quants Q6/Q8

Older Models - High Quality / Hard to Find

Older Models: Solar Models (Q6) - Exceptional Performance

Older Models: Mini-MOEs - Mixture of Experts 2x, x4 and x8

50 Leaderboards, Benchs, GGUF Tools, and Utilities

2000+ Run LLMs here - Directly in your browser

Rogue Creative Series - GGUF / Incl Source too.

updated 20 days ago

Based on Llama 3.2 this is a specialized creative series. L3.2 has strong instruction following. Repos including example generations. Source below too

DavidAU/L3.2-Rogue-Creative-Instruct-Uncensored-Abliterated-7B-GGUF

Text Generation • 8B • Updated Sep 25 • 26.2k • 110
DavidAU/Llama-3.2-9B-Uncensored-Brainstorm-Alpha-GGUF

Text Generation • 9B • Updated Jul 28 • 4.58k • 18
DavidAU/L3.2-Rogue-Creative-Instruct-Uncensored-7B-GGUF

Text Generation • 8B • Updated Jul 28 • 987 • 8
DavidAU/L3.2-Rogue-Creative-Instruct-Uncensored-Abliterated-7B

Text Generation • 8B • Updated Jul 28 • 1k • 7
DavidAU/L3.2-Rogue-Creative-Instruct-7B-GGUF

Text Generation • 7B • Updated Jul 28 • 649 • 4
DavidAU/L3.2-Rogue-Creative-Instruct-DARK-HORROR-Uncensored-Abliterated-7B

Text Generation • 7B • Updated Sep 28 • 43 • 3
DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters

Updated Jul 27 • 137

Note Document detailing all parameters, settings, samplers and advanced samplers to use not only my models to their maximum potential - but all models (and quants) online (regardless of the repo) to their maximum potential. Included quick start and detailed notes, include AI / LLM apps and other critical information and references too. A must read if you are using any AI/LLM right now.
DavidAU/AI_Autocorrect__Auto-Creative-Enhancement__Auto-Low-Quant-Optimization__gguf-exl2-hqq-SOFTWARE

Text Generation • Updated Jul 27 • 58

Note SOFTWARE patch (by me) for Silly Tavern (front end to connect to multiple AI apps / connect to AIs- like Koboldcpp, Lmstudio, Text Gen Web UI and other APIs) to control and improve output generation of ANY AI model. Also designed to control/wrangle some of my more "creative" models and make them perform perfectly with little to no parameter/samplers adjustments too.