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

HelloSun
/
SmallThinker4b

Text Generation
llama-cpp-python
GGUF
llama.cpp
Mixture of Experts
ssd-offload
smallthinker
expert-paging
low-ram
Model card Files Files and versions
xet
Community

Instructions to use HelloSun/SmallThinker4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

  • Libraries
  • llama-cpp-python

    How to use HelloSun/SmallThinker4b with llama-cpp-python:

    # !pip install llama-cpp-python
    
    from llama_cpp import Llama
    
    llm = Llama.from_pretrained(
    	repo_id="HelloSun/SmallThinker4b",
    	filename="{{GGUF_FILE}}",
    )
    
    output = llm(
    	"Once upon a time,",
    	max_tokens=512,
    	echo=True
    )
    print(output)
  • Notebooks
  • Google Colab
  • Kaggle
SmallThinker4b
135 kB
Ctrl+K
Ctrl+K
  • 2 contributors
History: 13 commits
Auto Upload Agent
v11: 方向修正(hot expert 實體放 RAM,非 page 轉換)+ b9acf138 vanilla 基線 82.1 tok/s / peak RSS 2.674 GiB
a035243 about 16 hours ago
  • patches
    v10: model card 加上完整實測數據(記憶體/速度/SSD/權重組成);修 pager expert 總量高估 7% 的 bug about 17 hours ago
  • tools
    v8: 修 io_probe 的 pread 計數 bug(SSD 337/667/1328/2558 MB/s)+ launcher arena 語意修正 about 17 hours ago
  • validate
    v11: 方向修正(hot expert 實體放 RAM,非 page 轉換)+ b9acf138 vanilla 基線 82.1 tok/s / peak RSS 2.674 GiB about 16 hours ago
  • .gitignore
    43 Bytes
    init about 18 hours ago
  • AGENTS.md
    4.59 kB
    v10: model card 加上完整實測數據(記憶體/速度/SSD/權重組成);修 pager expert 總量高估 7% 的 bug about 17 hours ago
  • README.md
    10.8 kB
    v10: model card 加上完整實測數據(記憶體/速度/SSD/權重組成);修 pager expert 總量高估 7% 的 bug about 17 hours ago
  • STATUS.md
    14.6 kB
    v11: 方向修正(hot expert 實體放 RAM,非 page 轉換)+ b9acf138 vanilla 基線 82.1 tok/s / peak RSS 2.674 GiB about 16 hours ago
  • llama_server.sh
    22.5 kB
    v8: 修 io_probe 的 pread 計數 bug(SSD 337/667/1328/2558 MB/s)+ launcher arena 語意修正 about 17 hours ago
  • sync.sh
    2.85 kB
    v0: 啟動器骨架 + 量測工具(未驗證) about 18 hours ago