David Belton's picture

In a Training Loop 🔄

David Belton PRO

DavidAU

·

AI & ML interests

Application(s) of single/multiple LLMs in specialized use cases & automation tasks. LLM, Prompt , System Role and Parameter engineering VIA chat / API. 2700+ merge models, over 600 fine tunes, over 1000 models published. DISCORD: David_AU [drawless111]

Recent Activity

new activity about 3 hours ago

DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking:MTP should be left in place, its trivial to add back in and it is beneficial.

new activity 1 day ago

DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking:Coaxing out nuanced personalities and sense of humor

new activity 1 day ago

DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking:Any updates to the Gemma model or are you fully switching to Qwen?

View all activity

Organizations

New activity in DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking about 3 hours ago

MTP should be left in place, its trivial to add back in and it is beneficial.

#4 opened about 6 hours ago by

New activity in DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking 1 day ago

Coaxing out nuanced personalities and sense of humor

#3 opened 3 days ago by

Any updates to the Gemma model or are you fully switching to Qwen?

#1 opened 5 days ago by

replied to their post 1 day ago

I will add to the list; may wait for specific Heretic and/or tuned version.

I already have a 43B-A3B version running in the lab ; however tuning these sparse moe models take a lot more work/time and ahh... detail. AND a lot more VRAM!!! [can't compress these atm, so BF16 required => 100 GB+ ]

updated 4 models 1 day ago

DavidAU/granite-4.1-8b-Brainstone-Thinking

Text Generation • 9B • Updated about 23 hours ago • 14

DavidAU/granite-4.1-8b-Brainstone2-Thinking

Text Generation • 9B • Updated about 23 hours ago • 6 • 1

DavidAU/granite-4.1-8b-Stone-Cold-Thinking

Text Generation • 9B • Updated 1 day ago • 3 • 1

DavidAU/granite-4.1-8b-FlintStones-V1

Text Generation • 9B • Updated 1 day ago • 60 • 1

published 2 models 1 day ago

DavidAU/granite-4.1-8b-Brainstone2-Thinking

Text Generation • 9B • Updated about 23 hours ago • 6 • 1

DavidAU/granite-4.1-8b-Brainstone-Thinking

Text Generation • 9B • Updated about 23 hours ago • 14

updated a model 2 days ago

DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking-NEO-CODE-Di-IMatrix-MAX-GGUF

Image-Text-to-Text • 39B • Updated 2 days ago • 33.6k • 17

published 2 models 3 days ago

DavidAU/granite-4.1-8b-Stone-Cold-Thinking

Text Generation • 9B • Updated 1 day ago • 3 • 1

DavidAU/granite-4.1-8b-FlintStones-V1

Text Generation • 9B • Updated 1 day ago • 60 • 1

updated 6 collections 3 days ago

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

Models that are dark/evil/corrupt (and many times NSFW!) to begin with then I add reasoning/thinking to them to make them even... ahh... better. • 121 items • Updated 3 days ago • 148

Grand Horror 16.5B+ Horror and Fiction Generation

NSFW, Horror, Fiction Generation models of different sizes and calibrations. Can be used for all genres of fiction. [ NEWEST -> bottom of the list ] • 146 items • Updated 3 days ago • 62

Heretic - Abliterated, Uncensored, Unrestricted POWER.

Models that have be abliterated using the HERETIC method. Done properly, this completely removed almost all censorship with no damage to the model. • 117 items • Updated 3 days ago • 74

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

Models (0.8B to 87B) in regular, "reasoning", "Brainstorm", MOE (1x to 8x / 128 experts), and expanded to create better and stronger code, faster. • 70 items • Updated 3 days ago • 29

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

Oldest models listed first, with Newest models at bottom of the page. Most repos have full examples, instructions, best settings and so on. • 273 items • Updated 3 days ago • 702

Qwen 3.5 - 0.8, 2, 4, 9, 27, 35B - regular / uncensored

Min 256k context + images : Reg, Heretic, Heretic fine tunes of Qwen 3.5 in all sizes. • 42 items • Updated 3 days ago • 41

published a model 3 days ago

DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking-NEO-CODE-Di-IMatrix-MAX-GGUF

Image-Text-to-Text • 39B • Updated 2 days ago • 33.6k • 17