16-bit Open on Hugging Face
Parameters
9.0B
Weights (est.)
≈ 18.0 GB
Mac memory (est.)
24 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/K2-Horizon-7B-chat-mlx-16bit --prompt "Hello"
Details
- Base model
- IFM/K2-Horizon-7B (finetune)
- Vendor
- IFM
- Task
- Text generation
- Quantisation
- 16-bit
- Licence
- apache-2.0
- Downloads
- 361 in the last 30 days · 361 all time · 0 likes
- Published
- 2026-10-14 · updated 2026-10-14 (28d ago)
- Tags
- k2_horizon, apple-silicon, k2-horizon, 7b, dense, open-weights, custom_code
Other MLX builds of IFM/K2-Horizon-7B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| K2-Horizon-7B-4bit | 4-bit | ≈ 5.1 GB | 8 GB+ | 555 |
| K2-Horizon-7B-oQ6e | 6-bit | ≈ 7.3 GB | 16 GB+ | 816 |
| K2-Horizon-7B-8bit | 8-bit | ≈ 9.6 GB | 16 GB+ | 148 |
| K2-Horizon-7B-chat-mlx-16bit (this one) | 16-bit | ≈ 18.0 GB | 24 GB+ | 361 |
| K2-Horizon-7B-bf16 | bf16 | ≈ 18.0 GB | 24 GB+ | 99 |