Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets
  • 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

hamuz
/
Nanbeige4.3-33B-MLX

Text Generation
MLX
Safetensors
English
Chinese
mlx-lm
apple-silicon
quantized
nanbeige
llama
agent
reasoning
tool-use
conversational
Model card Files Files and versions
xet
Community

Instructions to use hamuz/Nanbeige4.3-33B-MLX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

  • Libraries
  • MLX

    How to use hamuz/Nanbeige4.3-33B-MLX with MLX:

    # Make sure mlx-lm is installed
    # pip install --upgrade mlx-lm
    
    # Generate text with mlx-lm
    from mlx_lm import load, generate
    
    model, tokenizer = load("hamuz/Nanbeige4.3-33B-MLX")
    
    prompt = "Write a story about Einstein"
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, add_generation_prompt=True
    )
    
    text = generate(model, tokenizer, prompt=prompt, verbose=True)
  • Notebooks
  • Google Colab
  • Kaggle
  • Local Apps Settings
  • LM Studio
  • MLX LM

    How to use hamuz/Nanbeige4.3-33B-MLX with MLX LM:

    Generate or start a chat session
    # Install MLX LM
    uv tool install mlx-lm
    # Interactive chat REPL
    mlx_lm.chat --model "hamuz/Nanbeige4.3-33B-MLX"
    Run an OpenAI-compatible server
    # Install MLX LM
    uv tool install mlx-lm
    # Start the server
    mlx_lm.server --model "hamuz/Nanbeige4.3-33B-MLX"
    # Calling the OpenAI-compatible server with curl
    curl -X POST "http://localhost:8080/v1/chat/completions" \
       -H "Content-Type: application/json" \
       --data '{
         "model": "hamuz/Nanbeige4.3-33B-MLX",
         "messages": [
           {"role": "user", "content": "Hello"}
         ]
       }'
  • Atomic Chat
Nanbeige4.3-33B-MLX
195 GB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 10 commits
hamuz's picture
hamuz
Model card with measured perplexity & speed for all variants
3bcbec0 verified 22 days ago
  • 2bit
    Add 2bit MLX weights 22 days ago
  • 3bit
    Add 3bit MLX weights 22 days ago
  • 4bit
    Add 4bit MLX weights 22 days ago
  • 5bit
    Add 5bit MLX weights 22 days ago
  • 6bit
    Add 6bit MLX weights 22 days ago
  • 8bit
    Add 8bit MLX weights 22 days ago
  • bf16
    Add bf16 MLX weights 22 days ago
  • figures
    Add model card (benchmarks in progress) 22 days ago
  • results
    Model card with measured perplexity & speed for all variants 22 days ago
  • .gitattributes
    2.04 kB
    Add bf16 MLX weights 22 days ago
  • README.md
    10.4 kB
    Model card with measured perplexity & speed for all variants 22 days ago