what/llm/can/i/run
models/google-gemma-2-27b-it

Gemma 2 27B

Sources arena, livebench · Confidence B
params
27B
dense
bench sources
2
mid conf
context
max tokens
license
Gemma license
see source

Model meta

canonical
google/gemma-2-27b-it
parameters
27B
organization
Google
license
Gemma license
context
downloads

Benchmark breakdown

LiveBench
instruction following68.8
coding35.9
language34.0
overall46.2

Hardware fit matrix

by gpu tierQ2_K
10.1 GB
Q4_K_M
13.5 GB
Q5_K_M
16.9 GB
Q6_K
20.3 GB
Q8_0
27.0 GB
FP16
54.0 GB
G8 GB tier
rtx 3050 · 4060
G12 GB tier
rtx 3060 · 4070
OK
36 t/s
G16 GB tier
rtx 4080 · 4060 ti 16g
GREAT
54 t/s
OK
41 t/s
G24 GB tier
rtx 3090 · 4090
GREAT
87 t/s
GREAT
66 t/s
GREAT
53 t/s
OK
45 t/s
G32 GB tier
rtx 5090 · m3 max
GREAT
162 t/s
GREAT
123 t/s
GREAT
99 t/s
GREAT
83 t/s
OK
62 t/s
G48 GB tier
a6000 · m3 max 64
GREAT
52 t/s
GREAT
40 t/s
GREAT
32 t/s
GREAT
27 t/s
GREAT
20 t/s
G80 GB tier
h100 · m3 ultra 128
GREAT
177 t/s
GREAT
134 t/s
GREAT
108 t/s
GREAT
91 t/s
GREAT
68 t/s
GREAT
34 t/s
full fit · production speed tight fit · usabledoesn't fit
Get personalized ranking
Tell us your machine — we'll tell you if this is actually your best pick, or what's better.
Rank for my hardware