GGUF models by logihertzsystems
40 GGUF models quantized by logihertzsystems on Hugging Face. Sizes run from 3.09B to 7.62B parameters. The longest trained context is 125K tokens.
About logihertzsystems on Hugging Face
logihertzsystems has published 40 GGUF models that ggufmodels.com tracks, across families such as Nyra. The most downloaded are Nyra Code 3B GGUF IQ1 M, Nyra Code 3B GGUF IQ1 S and Nyra Code 3B GGUF Q4 K M.
The quants logihertzsystems publishes most often are Q8_0 (10), Q4_K_M (10), IQ1_S (10) and IQ1_M (10). Their Q4_K_M files range from 1.9 to 4.7 GB. Parameter counts run from 3.09B to 7.62B, and the longest trained context window is 125K tokens.
Frequently asked questions
What GGUF models has logihertzsystems published?
40 models are listed here, mainly from the Nyra families. The most downloaded is Nyra Code 3B GGUF IQ1 M.
Which quants does logihertzsystems publish?
The most common are Q8_0 (10 models), Q4_K_M (10 models), IQ1_S (10 models) and IQ1_M (10 models).