TEST #1968 RESULTS
12/19/2025 - 9:19 PM
ACCELERATOR
MODEL
2.0
tokens/s
generation
16.03
sec
time to first token
110
tokens/s
prompt
24
LocalScore
HOW YOU STACK UP
Explore All ResultsMistral Small 22B ArliAI RPMax v1.1 - Q5_K - Medium
SYSTEM
CPU
AMD Ryzen 5 5600GT with Radeon Graphics (znver3)
RAM
31.8GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
81
tokens/s
2.0
tokens/s
13.12
sec
pp4096+tg256
50
tokens/s
1.2
tokens/s
82.40
sec
pp2048+tg256
156
tokens/s
2.1
tokens/s
13.56
sec
pp2048+tg768
163
tokens/s
2.1
tokens/s
13.03
sec
pp1024+tg1024
162
tokens/s
2.2
tokens/s
6.77
sec
pp1280+tg3072
142
tokens/s
1.9
tokens/s
9.55
sec
pp384+tg1152
126
tokens/s
2.2
tokens/s
3.57
sec
pp64+tg1024
86
tokens/s
2.2
tokens/s
1.18
sec
pp16+tg1536
26
tokens/s
2.2
tokens/s
1.07
sec