Bloom GGUF models
151 GGUF models in the Bloom family, quantized by 18 uploaders. Sizes run from 34M to 180B parameters. The longest trained context is 32K tokens.
About the Bloom GGUF models
Bloom is represented on ggufmodels.com by 151 GGUF models, quantized by 18 uploaders on Hugging Face. The most downloaded are bloom 560m, Bloom 1b7 creative writing Cont IT Step5 i1 and Bloom 560m. Parameter counts run from 34M to 180B, and the longest trained context window is 32K tokens.
A Q4_K_M file of a Bloom model is between 0.4 to 114.7 GB, which is the quant most people start with. Most files come from RichardErkhov (76), mradermacher (42) and tensorblock (8). The quants you will meet most often are Q2_K (132), Q3_K_S (123), Q3_K_M (119) and Q4_K_M (116), where the number is how many models offer that quant.
Frequently asked questions
Which Bloom GGUF should I download?
The most downloaded Bloom GGUF here is bloom 560m. 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 Bloom model need?
Q4_K_M files of Bloom models on this site range from 0.4 to 114.7 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 Bloom GGUF files come from?
The most common quantizers here are RichardErkhov, mradermacher and tensorblock. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.