Large GGUF models
15744 GGUF models with 35B parameters and up. Sizes run from 34.1B to 20260731B parameters. The longest trained context is 10240K tokens.
What counts as a large model?
On this site large means 35B parameters and up. A Q4_K_M file is usually 22 GB or more, so you need a 24 GB or larger GPU, several GPUs or a lot of system RAM. 15,744 models fit this class. The most downloaded are Qwen3.8 Flash Next GSQ RCO, Ornith-1.5-35B and Qwen3.8 Flash Next GSQ RCO abliterated.
Their Q4_K_M files run 0.0 to 1481.4 GB. The families you will see most are Qwen (3189), Llama (2784) and DeepSeek (761). The size class sorts models by parameter count, so a model with more experts or a longer context can need more memory than its class suggests. The model pages show the real file sizes.
Frequently asked questions
How much VRAM does a large GGUF model need?
A Q4_K_M file is usually 22 GB or more, so you need a 24 GB or larger GPU, several GPUs or a lot of system RAM.
Which large GGUF model is the most popular?
By Hugging Face downloads, Qwen3.8 Flash Next GSQ RCO.