BGE GGUF models
296 GGUF models in the BGE family, quantized by 142 uploaders. Sizes run from 17M to 30.5B parameters. The longest trained context is 125K tokens.
About the BGE GGUF models
BGE is represented on ggufmodels.com by 296 GGUF models, quantized by 142 uploaders on Hugging Face. The most downloaded are bge small en v1.5, bge-small-en-v1.5 and bge-large-en-v1.5. Parameter counts run from 17M to 30.5B, and the longest trained context window is 125K tokens.
A Q4_K_M file of a BGE model is between 0.0 to 5.0 GB, which is the quant most people start with. Most files come from keisuke-miyako (42), mradermacher (39) and sinjab (9). The quants you will meet most often are Q8_0 (177), Q4_K_M (133), Q5_K_M (70) and F16 (66), where the number is how many models offer that quant.
Frequently asked questions
Which BGE GGUF should I download?
The most downloaded BGE GGUF here is bge small en v1.5. 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 BGE model need?
Q4_K_M files of BGE models on this site range from 0.0 to 5.0 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 BGE GGUF files come from?
The most common quantizers here are keisuke-miyako, mradermacher and sinjab. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.