K2-Horizon-375B-A23B: hardware requirements
K2-Horizon-375B-A23B needs about 233 GB of memory at Q4 quantization — a 229 GB download plus room for context and overhead. That is beyond consumer hardware; use it through an API instead.
This is a mixture-of-experts model. Only a fraction of its 379B parameters are used for each token, so it runs faster than its size suggests — but all of it still has to fit in memory. The speed is that of a small model; the memory requirement is that of a large one. More on this →
Will it run on your machine?
Download sizes by quantization
Lower quantization means a smaller file and less memory, at some cost to quality. Q4_K_M is the usual starting point. What is this?
| Quant | Download | RAM @ 4K | RAM @ 32K | Source |
|---|---|---|---|---|
| Q4_K_Mrecommended | 229 GB | 233 GB | 240 GB | estimated |
| Q5_K_M | 269 GB | 273 GB | 280 GB | estimated |
| Q6_K | 311 GB | 315 GB | 322 GB | estimated |
| Q8_0 | 403 GB | 407 GB | 414 GB | estimated |
| BF16 | 758 GB | 763 GB | 770 GB | estimated |
“Measured” sizes are read from published GGUF files. “Estimated” sizes are derived from the parameter count and are typically within a few percent. The RAM columns add the context cache and runtime overhead to the download size.