Instructions to use drowzeys/keys-Mac-DeepSeek-V4-Flash-0731-MXFP4-MLX-Abliterated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use drowzeys/keys-Mac-DeepSeek-V4-Flash-0731-MXFP4-MLX-Abliterated 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("drowzeys/keys-Mac-DeepSeek-V4-Flash-0731-MXFP4-MLX-Abliterated") 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 drowzeys/keys-Mac-DeepSeek-V4-Flash-0731-MXFP4-MLX-Abliterated with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Generate some text mlx_lm.generate --model "drowzeys/keys-Mac-DeepSeek-V4-Flash-0731-MXFP4-MLX-Abliterated" --prompt "Once upon a time"
- Atomic Chat
Drowzeys
Edited - Thanks keys..
Sorry. Access approved.. i thought i set it to approve … had doing a lot of work on the dgx sparks..
Thanks..apologies for the rant. I genuinely thought you were just ignoring me. I'll know for future reference.
Don’t worry bro… i got you… l thought i had it turned up for agent review and approve.. didn’t hook it up correctly. Its fixed now.. without this huggingface weights here if you just do the github, it’ll just download the stock version which explains why you got the regular model. I’m upgrading the recipe soon hopefully will run even faster… thank you for being patient and message me.