MLX Community · First MLX build 25 Jun 2025
Llama-3.2-8X4B-MOE-V2-Dark-Champion-Instruct-uncensored-abliterated-21B-Q_6
MLX
Text
A 21B parameter text generation model, with one MLX build at 6-bit.
- Parameters
- 21B From Hugging Face
- Quantisation
- 6-bit Llama-3.2-8X4B-MOE-V2-Dark-Champion-Instruct-uncensored-abliterated-21B-Q_6-MLX
- Memory needed
- 21 GB Estimate, unified memory
- Download
- 17.0 GB Weights, estimated
- Downloads
- 2.5K This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 1 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 21 GB at 6-bit
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.
Run it
Terminal · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Llama-3.2-8X4B-MOE-V2-Dark-Champion-Instruct-uncensored-abliterated-21B-Q_6-MLX --prompt "Hello"MLX builds
1 on mlx-community| Llama-3.2-8X4B-MOE-V2-Dark-Champion-Instruct-uncensored-abliterated-21B-Q_6-MLX | 6-bit | 17.0 GB | 21 GB | 2.5K |
|---|
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.