PrimeIntellect · First MLX build 27 Nov 2025
INTELLECT-3
MLX
Text mit
A 107B parameter text generation model from PrimeIntellect, with 5 MLX builds, from 4-bit to bf16.
- Parameters
- 107B From Hugging Face
- Quantisation
- 4-bit INTELLECT-3-4bit
- Memory needed
- 70 GB Estimate, unified memory
- Download
- 60.1 GB Weights, estimated
- Downloads
- 129 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 5 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 70 GB at 4-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/INTELLECT-3-4bit --prompt "Hello"MLX builds
5 on mlx-community| INTELLECT-3-4bit | 4-bit | 60.1 GB | 70 GB | 129 |
|---|---|---|---|---|
| INTELLECT-3-5bit | 5-bit | 73.5 GB | 85 GB | 28 |
| INTELLECT-3-6bit | 6-bit | 86.8 GB | 100 GB | 29 |
| INTELLECT-3-8bit | 8-bit | 114 GB | 129 GB | 25 |
| INTELLECT-3-bf16 | bf16 | 214 GB | 241 GB | 30 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.