Tiny GGUF models
150 GGUF models in the Tiny family, quantized by 51 uploaders. Sizes run from 3M to 3.88B parameters. The longest trained context is 488K tokens.
About the Tiny GGUF models
Tiny is represented on ggufmodels.com by 150 GGUF models, quantized by 51 uploaders on Hugging Face. The most downloaded are Tiny Aya Global, Tiny LLM and Tiny Pirate 1.1b v0.2 i1. Parameter counts run from 3M to 3.88B, and the longest trained context window is 488K tokens.
A Q4_K_M file of a Tiny model is between 0.0 to 2.5 GB, which is the quant most people start with. Most files come from mradermacher (57), tensorblock (9) and RichardErkhov (8). The quants you will meet most often are Q4_K_M (97), Q2_K (88), Q5_K_M (82) and Q8_0 (79), where the number is how many models offer that quant.
Frequently asked questions
Which Tiny GGUF should I download?
The most downloaded Tiny GGUF here is Tiny Aya Global. 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 Tiny model need?
Q4_K_M files of Tiny models on this site range from 0.0 to 2.5 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 Tiny GGUF files come from?
The most common quantizers here are mradermacher, tensorblock and RichardErkhov. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.