Loading…
Loading…
Models / Gemma-4
1 model (12B) · ↓ 36.6K downloads a month · from $0.19/h on 2x 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 |
|---|---|---|---|---|
| OriginalAEON-7/Gemma-4-12B-it-AEON-Abliterated-K4-BF16 | vLLM | 24 GB | 32.4 GB | $0.42/h 4x RTX 3060 |
| Gemma 4 12B it AEON Abliterated K4 NVFP4AEON-7/Gemma-4-12B-it-AEON-Abliterated-K4-NVFP4 | vLLM | 12 GB | 18.8 GB | $0.19/h 2x RTX 3060 |
| Gemma 4 12B it AEON Abliterated K4 FP8AEON-7/Gemma-4-12B-it-AEON-Abliterated-K4-FP8 | vLLM | 13 GB | 20.3 GB | $0.19/h 2x RTX 3060 |
| Gemma 4 12B it AEON Abliterated K4 NVFP4 FP8AEON-7/Gemma-4-12B-it-AEON-Abliterated-K4-NVFP4-FP8 | vLLM | 9.3 GB | 16.1 GB | $0.19/h 2x 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").