Files and memory at 8K context
| Quant | File | Files | + KV + runtime | Smallest fit | Source |
|---|---|---|---|---|---|
| Q4_K_M IFM/K2-Horizon-0.9B-GGUF | 0.62 GiB | 1 | 2.1 GiB | 8 GB GPU | HF files |
| Q5_K_M IFM/K2-Horizon-0.9B-GGUF | 0.72 GiB | 1 | 2.2 GiB | 8 GB GPU | HF files |
| Q8_0 IFM/K2-Horizon-0.9B-GGUF | 1.07 GiB | 1 | 2.5 GiB | 8 GB GPU | HF files |
| FP16 mradermacher/K2-Horizon-0.9B-GGUF | 2.01 GiB | 1 | 3.4 GiB | 8 GB GPU | HF files |
| MLX 4-bit mixed precision mlx-community/K2-Horizon-0.9B-4bit | 0.80 GiB | 1 | 2.2 GiB | 8 GB Mac | HF files |
| MLX 8-bit mlx-community/K2-Horizon-0.9B-8bit | 1.07 GiB | 1 | 2.5 GiB | 8 GB Mac | HF files |
Sizes are exact file bytes (text weights; projectors listed separately). “+” = recurrent state or other terms unknown. Smallest fit adds 0.5 GiB spare; Macs use 75% of memory.
Quality
Artificial Analysis Intelligence Index
Unknown
Arena text rating
Unknown
Scores describe the evaluated hosted variant, not each quantized file. AA and Arena are never averaged.
Reported speeds 0
No reported measurement yet. Submit one.
Community threads
- Hyperparameters Used During Training Are Not Documented Hugging Face · 2026-10-12
- K2-Horizon public release artifacts: four template and documentation observations Hugging Face · 2026-10-07
Qualitative signals only; titles are not verified claims.