Spark GGUF models
161 GGUF models in the Spark family, quantized by 73 uploaders. Sizes run from 218M to 8.9B parameters. The longest trained context is 1024K tokens.
About the Spark GGUF models
Spark is represented on ggufmodels.com by 161 GGUF models, quantized by 73 uploaders on Hugging Face. The most downloaded are Sharp Spark X2.5 4B, Spark X2.5 4B and Spark X2.5 4B. Parameter counts run from 218M to 8.9B, and the longest trained context window is 1024K tokens.
A Q4_K_M file of a Spark model is between 0.3 to 3.5 GB, which is the quant most people start with. Most files come from mradermacher (56), darioooooo0o (4) and XHToken (4). The quants you will meet most often are Q4_K_M (123), Q8_0 (115), Q6_K (104) and Q5_K_M (93), where the number is how many models offer that quant.
Frequently asked questions
Which Spark GGUF should I download?
The most downloaded Spark GGUF here is Sharp Spark X2.5 4B. 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 Spark model need?
Q4_K_M files of Spark models on this site range from 0.3 to 3.5 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 Spark GGUF files come from?
The most common quantizers here are mradermacher, darioooooo0o and XHToken. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.