IFM · First MLX build 4 Sept 2026
K2-Horizon-7B
MLX
Text apache-2.0
A 9B parameter text generation model from IFM, with 5 MLX builds, from 4-bit to bf16.
- Parameters
- 9B From Hugging Face
- Quantisation
- 6-bit K2-Horizon-7B-oQ6e
- Memory needed
- 11 GB Estimate, unified memory
- Download
- 7.3 GB Weights, estimated
- Downloads
- 794 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 11 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/K2-Horizon-7B-oQ6e --prompt "Hello"MLX builds
5 on mlx-community| K2-Horizon-7B-4bit | 4-bit | 5.1 GB | 8 GB | 545 |
|---|---|---|---|---|
| K2-Horizon-7B-oQ6e | 6-bit | 7.3 GB | 11 GB | 794 |
| K2-Horizon-7B-8bit | 8-bit | 9.6 GB | 13 GB | 150 |
| K2-Horizon-7B-chat-mlx-16bit | 16-bit | 18.0 GB | 23 GB | 366 |
| K2-Horizon-7B-bf16 | bf16 | 18.0 GB | 23 GB | 101 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.