TII · First MLX build 10 Jun 2025
Falcon3-7B-Instruct-1.58bit
MLX
Text falcon-llm-license
A 3.5B parameter text generation model from TII, with 4 MLX builds at 58-bit.
- Parameters
- 3.5B From Hugging Face
- Quantisation
- 58-bit Falcon3-7B-Instruct-1.58bit-4bit
- Memory needed
- 7 GB Estimate, unified memory
- Download
- 3.8 GB Weights, estimated
- Downloads
- 54 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 4 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 7 GB at 58-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/Falcon3-7B-Instruct-1.58bit-4bit --prompt "Hello"MLX builds
4 on mlx-community| Falcon3-7B-Instruct-1.58bit-6bit | 58-bit | 2.3 GB | 5 GB | 32 |
|---|---|---|---|---|
| Falcon3-7B-Instruct-1.58bit-8bit | 58-bit | 2.5 GB | 5 GB | 33 |
| Falcon3-7B-Instruct-1.58bit | 58-bit | 3.3 GB | 6 GB | 40 |
| Falcon3-7B-Instruct-1.58bit-4bit | 58-bit | 3.8 GB | 7 GB | 54 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.