second_constantine/
gemma3:27b

77 1 month ago

Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. (quantized UD-Q4_K_XL)

vision 27b

1 month ago

5d301428bccc · 18GB ·

model
gemma3
·
27B
·
Q4_K_M
projector
clip
·
423M
·
F16
template
{{- $systemPromptAdded := false }} {{- range $i, $_ := .Messages }} {{- $last := eq (len (slice $.Me
params
{ "min_p": 0.01, "num_predict": 32768, "repeat_penalty": 1, "stop": [ "<end_

Readme

Feature Value
vision true (>=0.11.11)
thinking false
tools false
Device Speed, token/s Context VRAM, gb Versions
RTX 3090 24gb ~34 4096 20 UD-Q4_K_XL,0.12.2
RTX 3090 24gb - 15360 “cudaMalloc failed” UD-Q4_K_XL,0.12.2
M1 Max 32gb ~13 4096 19 UD-Q4_K_XL,0.12.2