William Pieh

WTPieh's activity

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

New activity in PiehSoft/Qwen3.6-40B-Deckard-MTP about 16 hours ago

I can't get it to load with llama.cpp

6

#5 opened 10 days ago by

sa13ma

New activity in PiehSoft/Qwen3.6-40B-Deckard-MTP 12 days ago

Q8 quant

2

#4 opened 12 days ago by

chrisburri

New activity in DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking-NEO-CODE-Di-IMatrix-MAX-GGUF about 1 month ago

I am getting very good results with this model but its a bit slow, any chance for MTP?

32

#17 opened about 1 month ago by

bissli

New activity in PiehSoft/Qwen3.6-40B-Deckard-MTP about 1 month ago

Single-needle retrieval holds to 1M tokens (3.85x YaRN) + the llama.cpp context-cap workaround.

🚀 1

1

#3 opened about 1 month ago by

WTPieh

Strix Halo gfx1151 ROCmFP4 support

7

#2 opened about 2 months ago by

BebopVox

New activity in PiehSoft/Qwen3.6-40B-Deckard-MTP about 2 months ago

not new and first

🤯 1

1

#1 opened about 2 months ago by

Black6spdZ

New activity in DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking about 2 months ago

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

👍👀 3

12

#4 opened 3 months ago by

tcclaviger

New activity in DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking-NEO-CODE-Di-IMatrix-MAX-GGUF about 2 months ago

Cannot use MTP

5

#6 opened 2 months ago by

nickmok