SimonPu/
gemma-3:27b-it-qat_Q4_K_XL

25 1 month ago

The Google Gemma 3 models are multimodal—processing text and images—and feature a 128K context window with support for over 140 languages....

vision
... /
params
dc971284b880 · 127B
{
"min_p": 0.01,
"num_ctx": 32768,
"repeat_penalty": 1,
"stop": [
"<end_of_turn>"
],
"temperature": 0.1,
"top_k": 64,
"top_p": 0.95
}