RecursiveSI · First MLX build 24 Oct 2026
flint
MLX
Text apache-2.0 Trending
A 40B parameter text generation model from RecursiveSI, with 9 MLX builds, from 2-bit to 8-bit.
- Parameters
- 40B From Hugging Face
- Quantisation
- 4-bit flint-OptiQ-4bit
- Memory needed
- 28 GB Estimate, unified memory
- Download
- 22.4 GB Weights, estimated
- Downloads
- 2.0K This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 9 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 28 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/flint-OptiQ-4bit --prompt "Hello"MLX builds
9 on mlx-community| flint-oQ2e | 2-bit | 12.5 GB | 16 GB | 413 |
|---|---|---|---|---|
| flint-oQ2.7e | 2-bit | 12.5 GB | 16 GB | 399 |
| flint-oQ2.5e | 2-bit | 12.5 GB | 16 GB | 375 |
| flint-oQ3e | 3-bit | 17.4 GB | 22 GB | 436 |
| flint-OptiQ-4bit | 4-bit | 22.4 GB | 28 GB | 2.0K |
| flint-oQ4e | 4-bit | 22.4 GB | 28 GB | 1.2K |
| flint-oQ5e | 5-bit | 27.4 GB | 33 GB | 443 |
| flint-oQ6e | 6-bit | 32.4 GB | 39 GB | 572 |
| flint-oQ8e | 8-bit | 42.3 GB | 50 GB | 830 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.