Hathor GGUF models
92 GGUF models in the Hathor family, quantized by 16 uploaders. Sizes run from 7.24B to 9.24B parameters. The longest trained context is 1024K tokens.
About the Hathor GGUF models
Hathor is represented on ggufmodels.com by 92 GGUF models, quantized by 16 uploaders on Hugging Face. The most downloaded are Hathor_Tahsin-L3-8B-v0.85, Hathor v4.6 i1 and Hathor_Unstable-L3-8B-v0.3. Parameter counts run from 7.24B to 9.24B, and the longest trained context window is 1024K tokens.
A Q4_K_M file of a Hathor model is between 4.4 to 6.3 GB, which is the quant most people start with. Most files come from mradermacher (46), bartowski (13) and QuantFactory (10). The quants you will meet most often are Q4_K_M (85), Q6_K (80), Q5_K_M (79) and Q5_K_S (78), where the number is how many models offer that quant.
Frequently asked questions
Which Hathor GGUF should I download?
The most downloaded Hathor GGUF here is Hathor_Tahsin-L3-8B-v0.85. 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 Hathor model need?
Q4_K_M files of Hathor models on this site range from 4.4 to 6.3 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 Hathor GGUF files come from?
The most common quantizers here are mradermacher, bartowski and QuantFactory. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.