Loading…
Loading…
Models / Huihui-gemma-4
GPU memory needed (smallest version of each): Huihui-gemma-4 12B about 32.4 GB, Huihui-gemma-4 26B about 61.8 GB, Huihui-gemma-4 31B about 80.7 GB. The cheapest runs on 2x RTX 3090 from $0.48/h right now.
4 models (12B to 31B) · ↓ 5.6K downloads a month · from $0.48/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 |
|---|---|---|---|---|
| Originalhuihui-ai/Huihui-gemma-4-12B-it-abliterated | vLLM | 24 GB | 32.4 GB | $0.48/h 2x RTX 3090 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originalhuihui-ai/Huihui-gemma-4-12B-coder-fable5-composer2.5-v1-abliterated | vLLM | 24 GB | 32.4 GB | $0.48/h 2x RTX 3090 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originalhuihui-ai/Huihui-gemma-4-26B-A4B-it-abliterated | vLLM | 52 GB | 61.8 GB | $0.89/h Nvidia A800 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originalhuihui-ai/Huihui-gemma-4-31B-it-abliterated-v2 | vLLM | 63 GB | 80.7 GB | $0.97/h 4x 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").