Loading…
Loading…
Models / Gemma-3-R1984
GPU memory needed (smallest version of each): Gemma-3-R1984 12B about 32.9 GB, Gemma-3-R1984 27B about 68.1 GB. The cheapest runs on 4x RTX 3060 from $0.35/h right now.
2 models (12B to 27B) · ↓ 51 downloads a month · from $0.35/h on 4x 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 |
|---|---|---|---|---|
| OriginalVIDraft/Gemma-3-R1984-12B | vLLM | 24 GB | 32.9 GB | $0.35/h 4x RTX 3060 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| OriginalVIDraft/Gemma-3-R1984-27BLicence to accept | vLLM | 55 GB | ~68.1 GB | $0.99/h 2x Nvidia A100 |
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").