David

xdavxd's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a collection 7 days ago

MeditronFO

Collection

The first fully open medical specialist LLM, outperforms Medgemma on open-ended clinical evaluation • 9 items • Updated Jun 25 • 4

New activity in RedHatAI/gemma-4-12B-it-NVFP4 8 days ago

Recipe ideas

2

#2 opened 17 days ago by

xdavxd

updated a model 11 days ago

xdavxd/gemma-4-26B-A4B-it-qat-heretic-ega-NVFP4

Image-Text-to-Text • 26B • Updated 11 days ago • 39

published a model 11 days ago

xdavxd/gemma-4-26B-A4B-it-qat-heretic-ega-NVFP4

Image-Text-to-Text • 26B • Updated 11 days ago • 39

updated a model 11 days ago

xdavxd/gemma-4-26B-A4B-it-qat-heretic-ega

Image-Text-to-Text • 26B • Updated 11 days ago • 75

published a model 13 days ago

xdavxd/gemma-4-26B-A4B-it-qat-heretic-ega

Image-Text-to-Text • 26B • Updated 11 days ago • 75

updated 2 models 16 days ago

xdavxd/gemma-4-12B-it-heretic-v2-NVFP4

Image-Text-to-Text • 12B • Updated 16 days ago • 258

xdavxd/gemma-4-26B-A4B-it-heretic-NVFP4

Image-Text-to-Text • 26B • Updated 16 days ago • 40

New activity in RedHatAI/gemma-4-26B-A4B-it-NVFP4 16 days ago

Better generation defaults?

#9 opened about 2 months ago by

xdavxd

New activity in xdavxd/gemma-4-12B-it-heretic-v2-NVFP4 17 days ago

Updated on 2026-09-02 to fix audio transcription and bake FP8 KV scales

#2 opened 17 days ago by

xdavxd

updated a model 17 days ago

xdavxd/gemma-4-12B-it-heretic-NVFP4

Image-Text-to-Text • 12B • Updated 16 days ago • 314

New activity in TrevorJS/gemma-4-12B-it-uncensored 17 days ago

Audio transcription degrades under NVFP4 quantization (BF16 is fine)

6

#1 opened 22 days ago by

xdavxd

liked a model 20 days ago

basically-ai/Pebble-10M-Chat

Text Generation • 11.1M • Updated 18 days ago • 1.49k • 13

updated a collection 22 days ago

Heretic NVFP4

Collection

NVFP4 quantized versions of de-censored models for use with vLLM. • 5 items • Updated 22 days ago

published a model 22 days ago

xdavxd/gemma-4-12B-it-heretic-v2-NVFP4

Image-Text-to-Text • 12B • Updated 16 days ago • 258

updated a collection 23 days ago

Heretic NVFP4

Collection

NVFP4 quantized versions of de-censored models for use with vLLM. • 5 items • Updated 22 days ago

updated a model 23 days ago

xdavxd/gemma-4-26B-A4B-it-heretic-v2-NVFP4

Image-Text-to-Text • 14B • Updated 23 days ago • 304 • 2

upvoted an article 25 days ago

view article

Article

Granite 4.2 LLMs: How They're Built

• 

ibm-granite

•

25 days ago

• 120

New activity in mradermacher/model_requests about 1 month ago

https://huggingface.co/swiss-ai/Apertus-v1.5-8B

7

#2763 opened about 1 month ago by

xdavxd

https://huggingface.co/swiss-ai/Apertus-v1.5-70B

2

#2764 opened about 1 month ago by

xdavxd