Models / MiniCPM5
MiniCPM5 by openbmb
9 models (1B to 2B) · ↓ 3.5M downloads a month · from $0.08/h on GTX 1660 SUPER 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.
MiniCPM5 1B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalopenbmb/MiniCPM5-1B | vLLM | 2.2 GB | 4.8 GB | $0.10/h RTX 3060 |
| Main (Prince-1)Prince-1/MiniCPM5-1B | ONNX Runtime GenAI | 2.2 GB | 4.2 GB | $0.08/h GTX 1660 SUPER |
| Main (onnx-community)onnx-community/MiniCPM5-1B | ONNX Runtime GenAI | 2.2 GB | 4.2 GB | $0.08/h GTX 1660 SUPER |
MiniCPM5 1B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalopenbmb/MiniCPM5-1B-MLX | vLLM | 610 MB | 3.1 GB | $0.10/h RTX 3060 |
MiniCPM5 1B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalopenbmb/MiniCPM5-1B-Base | vLLM | 2.2 GB | 4.8 GB | $0.10/h RTX 3060 |
MiniCPM5 1B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalopenbmb/MiniCPM5-1B-SFT | vLLM | 2.2 GB | 4.8 GB | $0.10/h RTX 3060 |
MiniCPM5 2B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalopenbmb/MiniCPM5-2B | vLLM | 5.0 GB | 8.2 GB | $0.10/h RTX 3060 |
MiniCPM5 2B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalopenbmb/MiniCPM5-2B-DSpark | vLLM | 650 MB | 3 GB | $0.10/h RTX 3060 |
MiniCPM5 2B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalopenbmb/MiniCPM5-2B-MLX | vLLM | 1.4 GB | 4.2 GB | $0.10/h RTX 3060 |
MiniCPM5 2B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalopenbmb/MiniCPM5-2B-Base | vLLM | 5.0 GB | 8.2 GB | $0.10/h RTX 3060 |
MiniCPM5 2B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalopenbmb/MiniCPM5-2B-SFT | vLLM | 5.0 GB | 8.2 GB | $0.10/h RTX 3060 |
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").