GGUF models by devparagiri
50 GGUF models quantized by devparagiri on Hugging Face. Sizes run from 124M to 3.82B parameters. The longest trained context is 128K tokens.
About devparagiri on Hugging Face
devparagiri has published 50 GGUF models that ggufmodels.com tracks, across families such as Llama, Dialogpt and Phi. The most downloaded are test 123 20250801 013353, Automotive 20250816 105217 and flow mint 20250824 175011.
The quants devparagiri publishes most often are Q8_0 (45) and Q4_K_M (5). Their Q4_K_M files range from 0.1 to 2.4 GB. Parameter counts run from 124M to 3.82B, and the longest trained context window is 128K tokens.
Frequently asked questions
What GGUF models has devparagiri published?
50 models are listed here, mainly from the Llama, Dialogpt and Phi families. The most downloaded is test 123 20250801 013353.
Which quants does devparagiri publish?
The most common are Q8_0 (45 models) and Q4_K_M (5 models).