Google · 33B · 密集（Dense）
Gemma 4 flagship base model (official) 看看你的 GPU 或 Mac 跑不跑得動 Gemma 4 31B——最低 18.4 GB，建議 30.7 GB。
2026-04256K context
| 量化 | 位元 | VRAM | 品質 | 状態 |
|---|---|---|---|---|
| Q2_K | 2 | 11.1 GB | low | — |
| Q3_K_M | 3 | 15.3 GB | moderate | — |
| Q4_K_M | 4 | 17.4 GB | good | — |
| Q5_K_M | 5 | 21.6 GB | good | — |
| Q6_K | 6 | 25.9 GB | excellent | — |
| Q8_0 | 8 | 34.3 GB | excellent | — |
| F16 | 16 | 68.1 GB | lossless | — |
Can I run Gemma 4 31B locally?
- Can I run Gemma 4 31B locally?
- Gemma 4 31B needs about 18.4 GB of memory at a minimum and 30.7 GB recommended. Open this page to grade it against your GPU or Mac, then run it with runai, Ollama or LM Studio.
- How much VRAM does Gemma 4 31B need?
- At Q4_K_M, Gemma 4 31B uses about 17.4 GB of VRAM. Higher quants need more memory; lower quants fit tighter cards with a quality tradeoff.