MiniCPM GGUF models
637 GGUF models in the MiniCPM family, quantized by 271 uploaders. Sizes run from 4M to 34.7B parameters. The longest trained context is 512K tokens.
About the MiniCPM GGUF models
MiniCPM is represented on ggufmodels.com by 637 GGUF models, quantized by 271 uploaders on Hugging Face. The most downloaded are MiniCPM5 2B, MiniCPM V 4.7 35B A3B and MiniCPM5-1B-Claude-Opus-Fable5-V2-Thinking. Parameter counts run from 4M to 34.7B, and the longest trained context window is 512K tokens.
A Q4_K_M file of a MiniCPM model is between 0.0 to 22.6 GB, which is the quant most people start with. Most files come from mradermacher (132), openbmb (24) and tinyopsec (17). The quants you will meet most often are Q4_K_M (434), Q8_0 (340), Q5_K_M (301) and Q6_K (296), where the number is how many models offer that quant.
Frequently asked questions
Which MiniCPM GGUF should I download?
The most downloaded MiniCPM GGUF here is MiniCPM5 2B. 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 MiniCPM model need?
Q4_K_M files of MiniCPM models on this site range from 0.0 to 22.6 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 MiniCPM GGUF files come from?
The most common quantizers here are mradermacher, openbmb and tinyopsec. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.