Will it run?
Everything to pick, install and run open models on your own computer.
Open modelsWill it run?Setup guidesLicensesHardwareCheat sheetsProjectsCompare modelsGlossaryRoadmapInterview Q&AFree resources
Will it run?
Pick a model, tell the page how much RAM and video memory your computer has, and it estimates whether the model will run on the GPU, run slowly on the CPU, or not fit at all. The estimate is parameters times bytes per weight plus about 1.5 GB of overhead, so treat it as a planning number and leave some headroom for your other programs.
| Model | Needs (est.) | Result |
|---|---|---|
| Gemma 2 9B | 6.9 GB | Runs on CPU (slow) |
| Llama 3.1 70B | 43.5 GB | Too big for this PC |
| Llama 3.1 8B | 6.3 GB | Runs on CPU (slow) |
| Mistral 7B Instruct v0.3 | 5.7 GB | Runs on CPU (slow) |
| Qwen2.5 7B Instruct | 6.1 GB | Runs on CPU (slow) |
Estimate: parameters x bytes per weight + about 1.5 GB overhead. Real use depends on context length and software.

