Will this AI model fit on your graphics card?

Enter the model, the quantization and the context length — see the memory it needs and which card it runs on.

Your model

Model size billions of parameters Quantization Context length tokens Context memory per 1,000 tokens GB

Result

–GB total

16 GB card–

–

32 GB card–

–

128 GB card–

–

Details

Weights	–	Bits per weight	–
Context memory	–	Overhead	1.5 GB
Total	–	Fits when total ≤	14.4 / 28.8 / 115.2 GB