Gemma 3 1B
by Google · gemma-3 family
1B
parameters
text-generation multilingual summarization
Gemma 3 1B is Google's ultra-lightweight model, ideal for edge devices and resource-constrained environments. Text-only (no vision at this size), it handles basic text generation and summarization tasks with minimal hardware requirements. At under 2 GB of VRAM for Q8, this model runs on virtually any modern hardware including older GPUs and base-config Macs.
Quick Start with Ollama
ollama run 1b-it-q8_0 | Creator | |
| Parameters | 1B |
| Architecture | transformer-decoder |
| Context | 32K tokens |
| Released | Mar 12, 2025 |
| License | Gemma Terms of Use |
| Ollama | gemma3:1b |
Quantization Options
| Format | File Size | VRAM Required | Quality | Ollama Tag |
|---|---|---|---|---|
| Q4_K_M | 0.8 GB | 1.5 GB |
★
★
★
★
★
| 1b-it-q4_K_M |
| Q8_0 rec | 1.1 GB | 2 GB |
★
★
★
★
★
| 1b-it-q8_0 |
| F16 | 2 GB | 3.5 GB |
★
★
★
★
★
| 1b-it-fp16 |
Compatible Hardware
for Q8_0 (2 GB VRAM)
Compatible Hardware
| Hardware | VRAM | Type | Fit | Est. Speed |
|---|---|---|---|---|
| Mac Studio M4 Ultra 512GB | 512 GB | mac | Runs | ~410 tok/s |
| Mac Pro M2 Ultra 192GB | 192 GB | mac | Runs | ~400 tok/s |
| Mac Studio M4 Ultra 192GB | 192 GB | mac | Runs | ~410 tok/s |
| Mac Studio M4 Max 128GB | 128 GB | mac | Runs | ~273 tok/s |
| MacBook Pro M4 Max 128GB | 128 GB | mac | Runs | ~273 tok/s |
| MacBook Pro M3 Max 96GB | 96 GB | mac | Runs | ~200 tok/s |
| Mac mini M4 Pro 64GB | 64 GB | mac | Runs | ~137 tok/s |
| Mac Studio M4 Max 64GB | 64 GB | mac | Runs | ~273 tok/s |
| MacBook Pro M4 Max 64GB | 64 GB | mac | Runs | ~273 tok/s |
| Mac mini M4 Pro 48GB | 48 GB | mac | Runs | ~137 tok/s |
| MacBook Pro M3 Max 48GB | 48 GB | mac | Runs | ~200 tok/s |
| MacBook Pro M4 Max 48GB | 48 GB | mac | Runs | ~273 tok/s |
| MacBook Pro M4 Pro 48GB | 48 GB | mac | Runs | ~137 tok/s |
| Mac Studio M4 Max 36GB | 36 GB | mac | Runs | ~273 tok/s |
| MacBook Pro M3 Pro 36GB | 36 GB | mac | Runs | ~75 tok/s |
| NVIDIA GeForce RTX 5090 | 32 GB | gpu | Runs | ~896 tok/s |
| iMac M4 32GB | 32 GB | mac | Runs | ~60 tok/s |
| Mac mini M4 32GB | 32 GB | mac | Runs | ~60 tok/s |
| MacBook Air M4 32GB | 32 GB | mac | Runs | ~60 tok/s |
| AMD Radeon RX 7900 XTX | 24 GB | gpu | Runs | ~480 tok/s |
| NVIDIA GeForce RTX 3090 | 24 GB | gpu | Runs | ~468 tok/s |
| NVIDIA GeForce RTX 4090 | 24 GB | gpu | Runs | ~504 tok/s |
| iMac M3 24GB | 24 GB | mac | Runs | ~50 tok/s |
| Mac mini M2 24GB | 24 GB | mac | Runs | ~50 tok/s |
| Mac mini M4 Pro 24GB | 24 GB | mac | Runs | ~137 tok/s |
| MacBook Air M2 24GB | 24 GB | mac | Runs | ~50 tok/s |
| MacBook Air M4 24GB | 24 GB | mac | Runs | ~60 tok/s |
| MacBook Pro M4 Pro 24GB | 24 GB | mac | Runs | ~137 tok/s |
| AMD Radeon RX 7900 XT | 20 GB | gpu | Runs | ~400 tok/s |
| MacBook Pro M3 Pro 18GB | 18 GB | mac | Runs | ~75 tok/s |
| AMD Radeon RX 6800 XT | 16 GB | gpu | Runs | ~256 tok/s |
| AMD Radeon RX 7800 XT | 16 GB | gpu | Runs | ~312 tok/s |
| Intel Arc A770 | 16 GB | gpu | Runs | ~280 tok/s |
| NVIDIA GeForce RTX 4060 Ti 16GB | 16 GB | gpu | Runs | ~144 tok/s |
| NVIDIA GeForce RTX 4070 Ti Super | 16 GB | gpu | Runs | ~336 tok/s |
| NVIDIA GeForce RTX 4080 Super | 16 GB | gpu | Runs | ~368 tok/s |
| NVIDIA GeForce RTX 4080 | 16 GB | gpu | Runs | ~359 tok/s |
| NVIDIA GeForce RTX 5070 Ti | 16 GB | gpu | Runs | ~448 tok/s |
| NVIDIA GeForce RTX 5080 | 16 GB | gpu | Runs | ~480 tok/s |
| iMac M1 16GB | 16 GB | mac | Runs | ~34 tok/s |
| iMac M4 16GB | 16 GB | mac | Runs | ~60 tok/s |
| Mac mini M1 16GB | 16 GB | mac | Runs | ~34 tok/s |
| Mac mini M4 16GB | 16 GB | mac | Runs | ~60 tok/s |
| MacBook Air M2 16GB | 16 GB | mac | Runs | ~50 tok/s |
| MacBook Air M3 16GB | 16 GB | mac | Runs | ~50 tok/s |
| MacBook Air M4 16GB | 16 GB | mac | Runs | ~60 tok/s |
| MacBook Pro M1 16GB | 16 GB | mac | Runs | ~34 tok/s |
| MacBook Pro M2 Pro 16GB | 16 GB | mac | Runs | ~100 tok/s |
| AMD Radeon RX 7700 XT | 12 GB | gpu | Runs | ~216 tok/s |
| NVIDIA GeForce RTX 3060 12GB | 12 GB | gpu | Runs | ~180 tok/s |
| NVIDIA GeForce RTX 3080 12GB | 12 GB | gpu | Runs | ~456 tok/s |
| NVIDIA GeForce RTX 4070 Super | 12 GB | gpu | Runs | ~252 tok/s |
| NVIDIA GeForce RTX 4070 Ti | 12 GB | gpu | Runs | ~252 tok/s |
| NVIDIA GeForce RTX 4070 | 12 GB | gpu | Runs | ~252 tok/s |
| NVIDIA GeForce RTX 5070 | 12 GB | gpu | Runs | ~336 tok/s |
| NVIDIA GeForce RTX 2080 Ti | 11 GB | gpu | Runs | ~308 tok/s |
| NVIDIA GeForce RTX 3080 10GB | 10 GB | gpu | Runs | ~380 tok/s |
| AMD Radeon RX 7600 | 8 GB | gpu | Runs | ~144 tok/s |
| Intel Arc A750 | 8 GB | gpu | Runs | ~256 tok/s |
| NVIDIA GeForce RTX 3060 Ti | 8 GB | gpu | Runs | ~224 tok/s |
| NVIDIA GeForce RTX 3070 | 8 GB | gpu | Runs | ~224 tok/s |
| NVIDIA GeForce RTX 4060 Ti 8GB | 8 GB | gpu | Runs | ~144 tok/s |
| NVIDIA GeForce RTX 4060 | 8 GB | gpu | Runs | ~136 tok/s |
| MacBook Air M1 8GB | 8 GB | mac | Runs | ~34 tok/s |
| MacBook Air M2 8GB | 8 GB | mac | Runs | ~50 tok/s |
Benchmark Scores
42.0
mmlu