Mirothinker GGUF models
78 GGUF models in the Mirothinker family, quantized by 14 uploaders. Sizes run from 4.02B to 235B parameters. The longest trained context is 256K tokens.
About the Mirothinker GGUF models
Mirothinker is represented on ggufmodels.com by 78 GGUF models, quantized by 14 uploaders on Hugging Face. The most downloaded are MiroThinker 1.7, MiroThinker 1.7 Mini and MiroThinker 1.7 Mini Heretic. Parameter counts run from 4.02B to 235B, and the longest trained context window is 256K tokens.
A Q4_K_M file of a Mirothinker model is between 2.5 to 142.2 GB, which is the quant most people start with. Most files come from mradermacher (48), DevQuasar (11) and bartowski (3). The quants you will meet most often are Q4_K_M (71), Q3_K_M (66), Q5_K_M (65) and Q2_K (65), where the number is how many models offer that quant.
Frequently asked questions
Which Mirothinker GGUF should I download?
The most downloaded Mirothinker GGUF here is MiroThinker 1.7. 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 Mirothinker model need?
Q4_K_M files of Mirothinker models on this site range from 2.5 to 142.2 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 Mirothinker GGUF files come from?
The most common quantizers here are mradermacher, DevQuasar and bartowski. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.