Ernie GGUF models
118 GGUF models in the Ernie family, quantized by 52 uploaders. Sizes run from 11M to 300B parameters. The longest trained context is 256K tokens.
About the Ernie GGUF models
Ernie is represented on ggufmodels.com by 118 GGUF models, quantized by 52 uploaders on Hugging Face. The most downloaded are ERNIE Image Turbo, Ernie-4.5-21B-A3B-Pt and Ernie-4.5-21B-A3B-Thinking. Parameter counts run from 11M to 300B, and the longest trained context window is 256K tokens.
A Q4_K_M file of a Ernie model is between 0.2 to 181.8 GB, which is the quant most people start with. Most files come from mradermacher (28), ERNIE-Community (11) and unsloth (6). The quants you will meet most often are Q4_K_M (76), Q6_K (60), Q8_0 (59) and Q3_K_M (59), where the number is how many models offer that quant.
Frequently asked questions
Which Ernie GGUF should I download?
The most downloaded Ernie GGUF here is ERNIE Image Turbo. 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 Ernie model need?
Q4_K_M files of Ernie models on this site range from 0.2 to 181.8 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 Ernie GGUF files come from?
The most common quantizers here are mradermacher, ERNIE-Community and unsloth. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.