Instructions to use mlx-community/DeepSeek-V3.2-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/DeepSeek-V3.2-4bit with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # if on a CUDA device, also pip install mlx[cuda] # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("mlx-community/DeepSeek-V3.2-4bit") prompt = "Once upon a time in" text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- MLX LM
How to use mlx-community/DeepSeek-V3.2-4bit with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Generate some text mlx_lm.generate --model "mlx-community/DeepSeek-V3.2-4bit" --prompt "Once upon a time"
Won't load in LM Studio
#2
by DiscoStew - opened
I was able to get https://huggingface.co/mlx-community/DeepSeek-V3.2-4bit working by removing "tokenizer_type": "deepseek_v32" from tokenizer_config.json. I also changed the Jinja template and uploaded it here: https://lmstudio.ai/discostew/deepseek-v3.2 I’ll file a bug with the issue loading it with mlx-lm 0.30.1 https://github.com/lmstudio-ai/lmstudio-bug-tracker/issues/1352