GGUF models by DevQuasar-8
86 GGUF models quantized by DevQuasar-8 on Hugging Face. Sizes run from 362M to 671B parameters. The longest trained context is 1000K tokens.
About DevQuasar-8 on Hugging Face
DevQuasar-8 has published 86 GGUF models that ggufmodels.com tracks, across families such as Qwen, Llama and Granite. The most downloaded are Ibm Granite.Granite 3b Code Base 2k, Qwen.Qwen2.5 Coder 14B Instruct and Nexusflow.Starling LM 7B beta.
The quants DevQuasar-8 publishes most often are Q2_K (86), Q8_0 (85), Q6_K (85) and Q5_K_M (85). Their Q4_K_M files range from 0.3 to 42.5 GB. Parameter counts run from 362M to 671B, and the longest trained context window is 1000K tokens.
Frequently asked questions
What GGUF models has DevQuasar-8 published?
86 models are listed here, mainly from the Qwen, Llama and Granite families. The most downloaded is Ibm Granite.Granite 3b Code Base 2k.
Which quants does DevQuasar-8 publish?
The most common are Q2_K (86 models), Q8_0 (85 models), Q6_K (85 models) and Q5_K_M (85 models).