Horizon GGUF models
112 GGUF models in the Horizon family, quantized by 41 uploaders. Sizes run from 1.08B to 379B parameters. The longest trained context is 512K tokens.
About the Horizon GGUF models
Horizon is represented on ggufmodels.com by 112 GGUF models, quantized by 41 uploaders on Hugging Face. The most downloaded are Diplodocus preview 4, K2 Horizon 7B and K2 Horizon MoVA 36B A4B. Parameter counts run from 1.08B to 379B, and the longest trained context window is 512K tokens.
A Q4_K_M file of a Horizon model is between 0.7 to 237.8 GB, which is the quant most people start with. Most files come from mradermacher (38), bartowski (6) and IFM (5). The quants you will meet most often are Q4_K_M (82), Q6_K (79), Q5_K_M (75) and Q8_0 (68), where the number is how many models offer that quant.
Frequently asked questions
Which Horizon GGUF should I download?
The most downloaded Horizon GGUF here is Diplodocus preview 4. Pick the size that fits your GPU first, then choose a quant: Q4_K_M is the usual starting point, and Q5_K_M or Q6_K give more quality if you have memory to spare. The VRAM pages below list what fits each GPU.
How much VRAM does a Horizon model need?
Q4_K_M files of Horizon models on this site range from 0.7 to 237.8 GB. Add the KV cache for your context length and about 0.5 GB for the runtime, which usually means 1.5 to 3 GB on top of the file at an 8K context.
Where do Horizon GGUF files come from?
The most common quantizers here are mradermacher, bartowski and IFM. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.