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. 500+ LLMs graded.

Recent Activity

updated a model about 2 hours ago

DavidAU/Qwen3.5-2B-Polaris-HighIQ-Thinking-x4

published a model about 2 hours ago

DavidAU/Qwen3.5-2B-Polaris-HighIQ-Thinking-x4

updated a model about 3 hours ago

DavidAU/Qwen3.5-2B-Polaris-HighIQ-Thinking-x3

View all activity

Organizations

Posts 7

Post

2778

Gemma 3 27B - The record breaker (Heretic'ed (uncensored) ; then training on Unsloth):

arc_challenge,arc_easy,boolq,hellaswag,openbookqa,piqa, winogrande
0.661 ,0.816 ,0.878,0.763 ,0.464 ,0.808 ,0.762

For comparison:
Qwen3.5-27B-Text
qx86-hi 0.443,0.498,0.857,0.701,0.372,0.770,0.752

Trained on a HERETIC uncensored base too ;

DavidAU/Gemma3-27B-it-vl-Polaris-HI16-Heretic-Uncensored-INSTRUCT

Post

5127

The "ERNIE" 21B MOE Distill High Reasoning Fine Tune Invasion:

3 Ernie 21B-A3B MOE Models (64 experts) fine tuned with Unsloth using Gemini Pro 3, Claude 4.5 Opus, and GLM 4.7 Flash high reasoning datasets.

All benched, all exceeding org model specs too.

https://huggingface.co/DavidAU/models?search=ernie

Enjoy the freedom and added power.

Collections 34

View 34 collections

spaces 1

GGUF Model VRAM Calculator

Calculate VRAM requirements for LLM models

models 374

DavidAU/Qwen3.5-2B-Polaris-HighIQ-Thinking-x3

Image-Text-to-Text • 2B • Updated about 1 hour ago

DavidAU/Qwen3.5-2B-Polaris-HighIQ-Thinking-x4

Image-Text-to-Text • 2B • Updated about 2 hours ago

DavidAU/Qwen3.5-2B-Polaris-HighIQ-Thinking-x2

Image-Text-to-Text • 2B • Updated about 2 hours ago

DavidAU/Qwen3.5-2B-Polaris-HighIQ-Thinking-x1

Image-Text-to-Text • 2B • Updated about 2 hours ago

DavidAU/Qwen3.5-27B-Gemini3-Pro-High-Reasoning-Compact-Thinking

Image-Text-to-Text • 27B • Updated about 3 hours ago • 1.48k • 3

DavidAU/Qwen3.5-27B-HERETIC-Polaris-Advanced-Thinking-Alpha-uncensored

Image-Text-to-Text • 27B • Updated about 3 hours ago • 2 • 2

DavidAU/Qwen3.5-27B-GLM-4.7-Flash-Thinking-ALPHA

27B • Updated about 22 hours ago • 2 • 1

DavidAU/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored

Image-Text-to-Text • 33B • Updated 3 days ago • 101

DavidAU/gemma-3-12b-it-vl-GLM-4.7-Flash-INSTRUCT-Thinking-Hybrid-Heretic-Uncensored

Image-Text-to-Text • 13B • Updated 3 days ago • 39

DavidAU/Qwen3-32B-VL-GLM-4.7-Flash-HI16-Heretic-Uncensored-Thinking

Image-Text-to-Text • 33B • Updated 3 days ago • 57 • 1

View 374 models

datasets 0

None public yet