OpenAI · First MLX build 22 Apr 2026
privacy-filter
MLX
Text apache-2.0
A 1.4B parameter token classification model from OpenAI, with 8 MLX builds, from 4-bit to bf16.
- Parameters
- 1.4B From Hugging Face
- Quantisation
- bf16 openai-privacy-filter-bf16
- Memory needed
- 6 GB Estimate, unified memory
- Download
- 2.8 GB Weights, estimated
- Downloads
- 418 This build, last 30 days
- Task
- Token classification Text
- MLX builds
- 8 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 6 GB at bf16
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 · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/openai-privacy-filter-bf16MLX builds
8 on mlx-community| openai-privacy-filter-4bit | 4-bit | 790 MB | 3 GB | 402 |
|---|---|---|---|---|
| openai-privacy-filter-mxfp4 | MXFP4 | 790 MB | 3 GB | 189 |
| openai-privacy-filter-5bit | 5-bit | 965 MB | 4 GB | 165 |
| openai-privacy-filter-6bit | 6-bit | 1.1 GB | 4 GB | 167 |
| openai-privacy-filter-8bit | 8-bit | 1.5 GB | 4 GB | 262 |
| openai-privacy-filter-mxfp8 | MXFP8 | 1.5 GB | 4 GB | 221 |
| openai-privacy-filter-bf16 | bf16 | 2.8 GB | 6 GB | 418 |
| openai-privacy-filter-nvfp4 | Unstated | 5.6 GB | 9 GB | 193 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.