Loading…
Loading…
Models / K2-Horizon
5 models (0.9B to 375B) · ↓ 163.2K downloads a month · from $0.10/h on RTX 3060 right now. Each model's versions do the same job; quantized ones (AWQ, FP8, 4-bit) need less GPU memory and run on cheaper machines.
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| OriginalIFM/K2-Horizon-0.9B | vLLM | 2.2 GB | 5.1 GB | $0.10/h RTX 3060 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| OriginalIFM/K2-Horizon-3.7B | vLLM | 10 GB | 14.8 GB | $0.16/h RTX 4060 Ti |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| OriginalIFM/K2-Horizon-7B | vLLM | 18 GB | 23.6 GB | $0.24/h RTX 3090 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| OriginalIFM/K2-Horizon-375B-A23B | vLLM | 758 GB | 847.1 GB | $44.03/h 8x Nvidia H200 |
Model details come from Hugging Face and are refreshed daily; prices follow the live offers. GPU memory is GridPup's estimate: the weights, room for an 8,192-token conversation (chat models) and the runtime's own memory. Large models are split across the GPUs of one machine ("4x RTX 4090").