Shihan Qu

zenmagnets's activity

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

liked a model about 4 hours ago

zai-org/GLM-4.7-Flash

Text Generation • 31B • Updated about 9 hours ago • 15.2k • • 651

liked a model about 5 hours ago

nvidia/Qwen3-Next-80B-A3B-Thinking-NVFP4

Text Generation • Updated 7 days ago • 11.3k • 16

liked a model about 6 hours ago

nvidia/Qwen3-Next-80B-A3B-Instruct-NVFP4

Text Generation • Updated 7 days ago • 12.4k • 12

New activity in GadflyII/GLM-4.7-Flash-NVFP4 about 6 hours ago

Really appreciate that you ran performance comparison tests with BF16!

2

#2 opened about 15 hours ago by

zenmagnets

liked 3 models about 6 hours ago

Tengyunw/MiniMax-M2.1-NVFP4

Text Generation • 115B • Updated 13 days ago • 96 • 5

cyankiwi/GLM-4.7-REAP-218B-A32B-AWQ-4bit

Text Generation • 38B • Updated 4 days ago • 1.34k • 4

0xSero/GLM-4.7-REAP-218B-A32B-W4A16

Text Generation • 2B • Updated 5 days ago • 488 • 14

New activity in zai-org/GLM-4.7-Flash about 7 hours ago

Enormous KV-cache size?

👍 ➕ 3

9

#3 opened about 20 hours ago by

nephepritou

New activity in marksverdhei/GLM-4.7-Flash-FP8 about 14 hours ago

Performance comps with BF16?

1

#3 opened about 14 hours ago by

zenmagnets

liked a model about 14 hours ago

cyankiwi/GLM-4.7-Flash-AWQ-8bit

Text Generation • 9B • Updated about 15 hours ago • 477 • 3

liked 2 models about 15 hours ago

GadflyII/GLM-4.7-Flash-NVFP4

Text Generation • 18B • Updated about 15 hours ago • 2.27k • 16

marksverdhei/GLM-4.7-Flash-FP8

Text Generation • Updated about 4 hours ago • 785 • 10

liked a model about 16 hours ago

cyankiwi/GLM-4.7-Flash-AWQ-4bit

Text Generation • 6B • Updated about 17 hours ago • 4.39k • 7

New activity in cyankiwi/GLM-4.7-Flash-AWQ-4bit about 17 hours ago

Any plans for a 6bit or 8bit version?

1

#3 opened about 17 hours ago by

zenmagnets

New activity in marksverdhei/GLM-4.7-Flash-FP8 about 17 hours ago

If 8bit, why shaped like 16 bit

2

#2 opened about 17 hours ago by

zenmagnets

liked 3 models 1 day ago

nightmedia/gpt-oss-120b-heretic-v2-mxfp4-q8-hi-mlx

Text Generation • 117B • Updated Nov 20, 2025 • 575 • 7

hrktos-37/gpt-oss-120b-heretic

Text Generation • 117B • Updated 2 days ago • 11 • 1

TacGibs/MiniMaxM2.1-EXL3-3.04bpw

Text Generation • 44B • Updated 2 days ago • 5 • 1

upvoted a collection 6 days ago

Cerebras REAP

Collection

Sparse MoE models compressed using REAP (Router-weighted Expert Activation Pruning) method • 22 items • Updated 8 days ago • 85

New activity in nvidia/Qwen3-30B-A3B-NVFP4 about 2 months ago

6 months since intro of NVFP4, and it's basically still a myth

1

#4 opened about 2 months ago by

zenmagnets