We have not been able to test all models on all configurations. Some models, such as the Llama 3.2 models, don't even ship with newer versions of Understand and are only included for legacy builds of Understand
| Nvidia GPU WIndows | AMD GPU Windows | Intel iGPU Windows | Nvidia GPU Linux | Unified Memory Apple ARM | |
| Qwen3-0.6 | ~5 GB | ~5.1 GB | ~3.7 GB | ||
| Qwen3-1.7 | ~5.6 GB | ~5.8 GB | ~6 GB | ~5.8 GB | ~3.7 GB |
| Qwen3-4 | ~8.6 GB | ~8.1 GB | ~4.8 GB | ||
| Qwen3-8 | ~11.2 GB | ~8.8 GB | |||
| Granite4.0-1 | ~4.6 GB | ~2.7 GB | |||
| Granite4.0-micro | ~6.3 GB | ~6.7 GB | ~7.2 GB | ~4.1 GB | |
| Llama3.2-1 | ~3.3 GB | ~4.6 GB | ~2.2 GB | ||
| Llama3.2-3 | ~8.1 GB | ~8.7 GB | ~5 GB | ||
| Llama3.2-8 | ~11.6 GB | ~9.5 GB |