Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets new
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Hardware
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up
logic65 's Collections
Whittle: research and dev
Whittle: models you can run
Qwen3.8 Whittle family

Whittle: research and dev

updated 7 days ago

The compression ladder, recipes, adapters and unrepaired variants behind the family.

Upvote
2

  • logic65/Qwen3.8-Whittle-dev

    Updated 11 days ago • 1

  • logic65/Qwen3.8-Whittle-v40-16.8B

    Text Generation • 18B • Updated 12 days ago • 395 • 1

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

    Text Generation • 18B • Updated 12 days ago • 1.12k • 1

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

    Text Generation • 16B • Updated 12 days ago • 7.34k • 2

  • logic65/Qwen3.8-Whittle-48L-cut

    Text Generation • Updated 12 days ago • 1

  • logic65/Qwen3.8-Whittle-44L-cut

    Text Generation • Updated 12 days ago • 1

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

    27B • Updated about 23 hours ago • 19.7k • 103

    Note Sparse MoE from the dense 27B. 32/39 on the battery but needs post training; published as method and measurements.


  • logic65/Qwen3.8-Whittle-tri-14.7B

    Text Generation • 15B • Updated 12 days ago • 2.75k • 1

  • logic65/Qwen3.8-Whittle-16B

    Text Generation • 16B • Updated 12 days ago • 5.79k • 2
Upvote
2
  • Collection guide
  • Browse collections
Company
TOS Privacy About Careers
Website
Models Datasets Spaces Pricing Docs