Prototype GGUF models
301 GGUF models in the Prototype family, quantized by 8 uploaders. Sizes run from 738M to 125B parameters. The longest trained context is 1000K tokens.
About the Prototype GGUF models
Prototype is represented on ggufmodels.com by 301 GGUF models, quantized by 8 uploaders on Hugging Face. The most downloaded are Prototype X 12b, Phonetic Tibetan To English Translation and G4 Prototype01 12B v0.1. Parameter counts run from 738M to 125B, and the longest trained context window is 1000K tokens.
A Q4_K_M file of a Prototype model is between 0.5 to 7.5 GB, which is the quant most people start with. Most files come from bruhzair (277), mradermacher (17) and Novaciano (2). The quants you will meet most often are Q6_K (152), Q8_0 (92), Q4_K (62) and Q4_K_M (20), where the number is how many models offer that quant.
Frequently asked questions
Which Prototype GGUF should I download?
The most downloaded Prototype GGUF here is Prototype X 12b. 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 Prototype model need?
Q4_K_M files of Prototype models on this site range from 0.5 to 7.5 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 Prototype GGUF files come from?
The most common quantizers here are bruhzair, mradermacher and Novaciano. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.