← CanIRunLocal · All hardware · All models
Google · 0.74B parameters · embedding.
Ollama weight files: Q8_0 1.3 GB (embeddinggemma-2:740m-q8_0); Default 1.3 GB (embeddinggemma-2:740m).
Yes on 56 machines in this list. The smallest is M1 8GB: Q8_0, about 1.9 GB, roughly 42 tokens per second, via Ollama. ollama pull embeddinggemma-2:740m-q8_0
ollama pull embeddinggemma-2:740m-q8_0ollama pull embeddinggemma-2:740m-q8_0ollama pull embeddinggemma-2:740m-q8_0ollama pull embeddinggemma-2:740m-q8_0| Hardware | Runtime | Quant | Memory | Fit | Speed | Command |
|---|---|---|---|---|---|---|
| M1 8GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~42 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M2 8GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~62 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M3 8GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~62 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 3060 Ti 8GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~207 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 3070 8GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~207 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 4060 8GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~126 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RX 7600 8GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~133 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 3080 10GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~351 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M1 16GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~42 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M2 16GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~62 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M3 16GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~62 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M4 16GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~74 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 3060 12GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~166 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 4070 12GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~233 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 5070 12GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~310 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 4060 Ti 16GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~133 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 4070 Ti Super 16GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~310 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 4080 16GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~331 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 5080 16GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~443 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RX 7800 XT 16GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~288 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RX 9070 XT 16GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~295 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M2 24GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~62 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M3 24GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~62 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M4 24GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~74 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 3090 24GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~432 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 4090 24GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~465 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RX 7900 XTX 24GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~443 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M1 Pro 32GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~123 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M2 Pro 32GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~123 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M4 32GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~74 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M3 Pro 36GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~92 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M4 Max (32c GPU) 36GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~252 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M5 Max (32c GPU) 36GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~283 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| RTX 5090 32GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~827 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M4 Pro 48GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~168 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M4 Max (40c GPU) 48GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~336 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M5 Max (40c GPU) 48GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~378 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M1 Max 64GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~246 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M2 Max 64GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~246 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M3 Max 64GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~246 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M4 Max 64GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~336 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M4 Max (40c GPU) 64GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~336 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M5 Max (40c GPU) 64GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~378 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| DGX Spark 64GB (ASUS, Dell, HP, MSI, Acer, Gigabyte) | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~158 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M3 Ultra 96GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~504 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M5 Ultra 96GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~738 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M3 Max 128GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~246 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M4 Max 128GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~336 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M4 Max (40c GPU) 128GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~336 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M5 Max (40c GPU) 128GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~378 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte) | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~158 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| M2 Ultra 192GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~492 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M3 Ultra 256GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~504 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M5 Ultra (64c GPU) 256GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~738 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M5 Ultra (80c GPU) 256GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~738 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
| Mac Studio M3 Ultra (80c GPU) 512GB | Ollama | Q8_0 | 1.9 GB | fits in fast memory | ~504 tok/s | ollama pull embeddinggemma-2:740m-q8_0 |
Estimates, not benchmarks. Model sizes are ollama.com download sizes; memory needed adds headroom for ~8k tokens of context; speeds are rough figures from memory bandwidth. Data updated 2026-10-16.
A project by OnDevice Labs · llms.txt