Moonshine GGUF models
91 GGUF models in the Moonshine family, quantized by 11 uploaders. Sizes run from 27M to 266M parameters.
About the Moonshine GGUF models
Moonshine is represented on ggufmodels.com by 91 GGUF models, quantized by 11 uploaders on Hugging Face. The most downloaded are Moonshine Streaming Medium, Moonshine Streaming Small and Moonshine Streaming Tiny. Parameter counts run from 27M to 266M.
A Q4_K_M file of a Moonshine model is between 0.1 to 0.1 GB, which is the quant most people start with. Most files come from cstr (20), myflow-ai (17) and handy-computer (17). The quants you will meet most often are Q8_0 (69), F16 (38), F32 (36) and Q4_K (20), where the number is how many models offer that quant.
Frequently asked questions
Which Moonshine GGUF should I download?
The most downloaded Moonshine GGUF here is Moonshine Streaming Medium. 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 Moonshine model need?
Q4_K_M files of Moonshine models on this site range from 0.1 to 0.1 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 Moonshine GGUF files come from?
The most common quantizers here are cstr, myflow-ai and handy-computer. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.