Codegemma GGUF models
98 GGUF models in the Codegemma family, quantized by 43 uploaders. Sizes run from 2.51B to 8.54B parameters. The longest trained context is 8K tokens.
About the Codegemma GGUF models
Codegemma is represented on ggufmodels.com by 98 GGUF models, quantized by 43 uploaders on Hugging Face. The most downloaded are codegemma-7b-it, codegemma-1.1-2b and codegemma-2b. Parameter counts run from 2.51B to 8.54B, and the longest trained context window is 8K tokens.
A Q4_K_M file of a Codegemma model is between 0.7 to 5.3 GB, which is the quant most people start with. Most files come from mradermacher (11), RichardErkhov (9) and google (5). The quants you will meet most often are Q4_K_M (73), Q5_K_M (59), Q2_K (58) and Q8_0 (54), where the number is how many models offer that quant.
Frequently asked questions
Which Codegemma GGUF should I download?
The most downloaded Codegemma GGUF here is codegemma-7b-it. 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 Codegemma model need?
Q4_K_M files of Codegemma models on this site range from 0.7 to 5.3 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 Codegemma GGUF files come from?
The most common quantizers here are mradermacher, RichardErkhov and google. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.