Instructions to use WaveCut/deepseek-ai_DeepSeek-R1-Distill-Llama-70B_MLX-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use WaveCut/deepseek-ai_DeepSeek-R1-Distill-Llama-70B_MLX-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir deepseek-ai_DeepSeek-R1-Distill-Llama-70B_MLX-4bit WaveCut/deepseek-ai_DeepSeek-R1-Distill-Llama-70B_MLX-4bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- LM Studio
metadata
base_model: deepseek-ai/DeepSeek-R1-Distill-Llama-70B
tags:
- mlx
WaveCut/deepseek-ai_DeepSeek-R1-Distill-Llama-70B_MLX-4bit
The Model WaveCut/deepseek-ai_DeepSeek-R1-Distill-Llama-70B_MLX-4bit was converted to MLX format from deepseek-ai/DeepSeek-R1-Distill-Llama-70B using mlx-lm version 0.21.1.
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("WaveCut/deepseek-ai_DeepSeek-R1-Distill-Llama-70B_MLX-4bit")
prompt = "hello"
if tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, add_generation_prompt=True
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)