Gemma 3 Models — Hardware Requirements
20 Gemma 3 models from Google and the community, from the smallest that runs in 0.5 GB of VRAM at Q4 up to 27.4B parameters. Every row links to full quantization tables and GPU compatibility.
All Gemma 3 Models by Size
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| Gemma 3 270M IT | 268M | 1.2 GB | — | ||
| Gemma 3 270M | 268M | 1.2 GB | — | ||
| Gemma 3 1B IT | 1000M | 2.1 GB | 33K | ||
| Gemma 3 1B Pt | 1000M | 2.1 GB | — | ||
| Gemma 3n E2B IT Litert Lm | 2B | 3.1 GB | — | ||
| Gemma 3 4B T1 IT | 3.9B | 3.4 GB | 131K | ||
| Gemma 3n E4B IT Litert Lm | 4B | 4.8 GB | — | ||
| Gemma 3 4B IT | 4.3B | 5 GB | — | ||
| Gemma 3 4B Pt | 4.3B | 5 GB | — | ||
| Gemma 3n E2B IT | 5.4B | 5.9 GB | — | ||
| Gemma 3n E4B IT | 7.8B | 7.7 GB | — | ||
| Gemma 3 12B IT | 12.2B | 10.9 GB | 33K | ||
| Gemma 3 12B Pt | 12.2B | 10.9 GB | — | ||
| Gemma 3 27B IT | 27.4B | 21.7 GB | 131K |
Community variants and fine-tunes(5)
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| Gemma 3 270M IT Heretic | 268M | 0.5 GB | 33K | ||
| Gemma 3 1B IT Heretic Extreme Uncensored Abliterated | 1000M | 1.0 GB | 33K | ||
| Gemma3 1B CulturaViva ITA | 1000M | 1.0 GB | 33K | ||
| Gemma 3 12B IT Heretic v2 | 12.2B | 9.2 GB | 131K | ||
| Gemma 3 12B IT Heretic | 12.2B | 9.2 GB | 131K |
How Gemma 3 Compares — llmrun Composite Score
Gemma 3 27B IT is the highest-scoring Gemma 3 model with an llmrun Composite Score of 79.4 — #91 among 106 open models. The top proprietary model, Gemini 4 Carbon, scores 158.6. Click a model to see its full benchmark breakdown.
Gemini 4 Carbon · proprietary158.6
GPT 6 Astra · proprietary156.2
Claude Opus 5.5 · proprietary153.0
GPT 6.1 Sol · proprietary151.7
GPT 6.1 Terra · proprietary150.1
Kimi K3.1 Pro142.8
MiMo V2.6 Pro RL133.3
Kimi K3132.7
MiMo V2.6 Flash RL131.9
GLM 5.3130.1
DeepSeek V4.1 Flash129.4
DeepSeek V4 Pro 0813129.2
Gemma 3 27B IT79.4
Gemma 3 12B IT74.2
Gemma 3 4B IT64.0
Frequently Asked Questions
- How much VRAM do I need to run a Gemma 3 model?
- The smallest Gemma 3 model, Gemma 3 270M IT Heretic, runs from 0.5 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 0.4 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which Gemma 3 models can I run on a 16 GB GPU?
- 18 of 19 Gemma 3 models fit in 16 GB of VRAM at Q4_K_M, including Gemma 3 1B IT, Gemma 3 4B IT, Gemma 3 12B IT.
- What is the most popular Gemma 3 model to run locally?
- Gemma 3 27B IT is the most downloaded Gemma 3 model in local-friendly quantized formats. It runs from 21.7 GB of VRAM at Q4_K_M.
- How do Gemma 3 models score on benchmarks?
- Gemma 3 27B IT leads the family with an llmrun Composite Score of 79.4, ranking #91 among 106 open models, while the top proprietary model, Gemini 4 Carbon, scores 158.6. See the comparison chart above for the full standings.