Loading…
Loading…
Models / Smaug
2 models (34B to 72B) · ↓ 17.4K downloads a month · from $0.74/h on 4x 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 |
|---|---|---|---|---|
| Originalabacusai/Smaug-34B-v0.1 | vLLM | 69 GB | 80.9 GB | $0.74/h 4x RTX 3090 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originalabacusai/Smaug-72B-v0.1 | vLLM | 145 GB | 186.7 GB | $2.77/h 2x RTX PRO 6000 |
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").