Loading…
Loading…
Models / deepseek-moe
2 models (16B to 16B) · ↓ 32.9K downloads a month · from $0.49/h on 2x RTX 3090 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 |
|---|---|---|---|---|
| Originaldeepseek-ai/deepseek-moe-16b-chat | vLLM | 33 GB | 39.7 GB | $0.49/h 2x RTX 3090 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originaldeepseek-ai/deepseek-moe-16b-base | vLLM | 33 GB | 39.7 GB | $0.49/h 2x RTX 3090 |
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").