scima

scima's activity

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

New activity in scima/Qwen3.8-Whittle-MoE-27B-A17.8B-GGUF-all-quants about 21 hours ago

Quants

15

#1 opened 1 day ago by

logic65

updated a model about 24 hours ago

scima/Qwen3.8-Whittle-MoE-27B-A17.8B-GGUF-all-quants

Text Generation • 27B • Updated about 24 hours ago • 1.1k • 1

liked a model 1 day ago

scima/Qwen3.8-Whittle-MoE-27B-A17.8B-GGUF-all-quants

Text Generation • 27B • Updated about 24 hours ago • 1.1k • 1

published a model 1 day ago

scima/Qwen3.8-Whittle-MoE-27B-A17.8B-GGUF-all-quants

Text Generation • 27B • Updated about 24 hours ago • 1.1k • 1

liked 2 models 5 days ago

logic65/Qwen3.8-Whittle-tri-14.7B-chat

15B • Updated 7 days ago • 2.09k • 2

logic65/Qwen3.8-Whittle-tri-kd-lora

Updated 7 days ago • 24 • 1

upvoted 3 collections 5 days ago

Whittle: research and dev

Collection

The compression ladder, recipes, adapters and unrepaired variants behind the family. • 9 items • Updated 3 days ago • 1

Whittle: models you can run

Collection

Compressed Qwen3.8-27B variants with weights published and tested. Start here. • 1 item • Updated 3 days ago • 1

Qwen3.8 Whittle family

Collection

A 27B whittled on two 8GB GPUs: measured cuts, heals, and restorations. 48L = best untrained, Whittle-16B = best overall. • 8 items • Updated 7 days ago • 1

liked 6 models 5 days ago

logic65/Qwen3.8-Whittle-tri-14.7B

Text Generation • 15B • Updated 7 days ago • 2.64k • 1

logic65/Qwen3.8-p44w75-16.8B-unrepaired

Text Generation • 16B • Updated 7 days ago • 7.22k • 2

logic65/Qwen3.8-Whittle-44L-cut

Text Generation • Updated 7 days ago • 1

logic65/Qwen3.8-Whittle-48L-cut

Text Generation • Updated 7 days ago • 1

logic65/Qwen3.8-Whittle-w50-18.3B-unrepaired

Text Generation • 18B • Updated 7 days ago • 1.12k • 1

logic65/Qwen3.8-Whittle-v40-16.8B

Text Generation • 18B • Updated 7 days ago • 392 • 1

liked 2 models 6 days ago

DavidAU/Qwen3.5-9B-The-Defiant-Fable-Uncensored-Heretic-NEO-IMATRIX-MAX-MTP-GGUF

Image-Text-to-Text • 9B • Updated 1 day ago • 865k • 469

logic65/Qwen3.8-Whittle-MoE-27B-A17.8B

27B • Updated 1 day ago • 12.4k • 52

liked 2 models 7 days ago

logic65/Qwen3.8-Whittle-dev

Updated 6 days ago • 1

logic65/about

Updated 9 days ago • 1

liked a model 8 days ago

logic65/Qwen3.8-Whittle-16B

Text Generation • 16B • Updated 7 days ago • 5.2k • 1