Heretic GGUF models
136 GGUF models in the Heretic family, quantized by 10 uploaders. Sizes run from 268M to 299B parameters. The longest trained context is 1024K tokens.
About the Heretic GGUF models
Heretic is represented on ggufmodels.com by 136 GGUF models, quantized by 10 uploaders on Hugging Face. The most downloaded are Hy MT2 1.8B Heretic, Heretic_Aquif 3.5 Nano 1B and heretic_HelpingAI 3B Coder. Parameter counts run from 268M to 299B, and the longest trained context window is 1024K tokens.
A Q4_K_M file of a Heretic model is between 0.3 to 21.2 GB, which is the quant most people start with. Most files come from mradermacher (121), ineosnexus (4) and OpenIntelligenceNet (3). The quants you will meet most often are Q6_K (127), Q4_K_M (127), Q5_K_M (122) and IQ4_XS (122), where the number is how many models offer that quant.
Frequently asked questions
Which Heretic GGUF should I download?
The most downloaded Heretic GGUF here is Hy MT2 1.8B Heretic. Pick the size that fits your GPU first, then choose a quant: Q4_K_M is the usual starting point, and Q5_K_M or Q6_K give more quality if you have memory to spare. The VRAM pages below list what fits each GPU.
How much VRAM does a Heretic model need?
Q4_K_M files of Heretic models on this site range from 0.3 to 21.2 GB. Add the KV cache for your context length and about 0.5 GB for the runtime, which usually means 1.5 to 3 GB on top of the file at an 8K context.
Where do Heretic GGUF files come from?
The most common quantizers here are mradermacher, ineosnexus and OpenIntelligenceNet. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.