GGUF models by Melvin56
62 GGUF models quantized by Melvin56 on Hugging Face. Sizes run from 494M to 10.3B parameters. The longest trained context is 128K tokens.
About Melvin56 on Hugging Face
Melvin56 has published 62 GGUF models that ggufmodels.com tracks, across families such as Qwen, Llama and Hammer. The most downloaded are DeepSeek R1 Distill Qwen 7B Abliterated v2, Phi 4 Mini Instruct Abliterated and Qwen2.5 Coder 7B Instruct Abliterated.
The quants Melvin56 publishes most often are Q8_0 (36), Q4_K_M (33), Q2_K (28) and Q5_K_M (26). Their Q4_K_M files range from 0.4 to 6.2 GB. Parameter counts run from 494M to 10.3B, and the longest trained context window is 128K tokens.
Frequently asked questions
What GGUF models has Melvin56 published?
62 models are listed here, mainly from the Qwen, Llama and Hammer families. The most downloaded is DeepSeek R1 Distill Qwen 7B Abliterated v2.
Which quants does Melvin56 publish?
The most common are Q8_0 (36 models), Q4_K_M (33 models), Q2_K (28 models) and Q5_K_M (26 models).