TEST #2278 RESULTS
01/13/2026 - 8:21 PM
ACCELERATOR
MODEL
38.2
tokens/s
generation
9.83
sec
time to first token
144
tokens/s
prompt
82
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC-Rome Processor (znver2)
RAM
7.6GB
OS
Linux
Kernel Release
6.12.57+deb13-cloud-amd64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Debian 6.12.57-1 (2025-11-05)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
152
tokens/s
42.5
tokens/s
6.76
sec
pp4096+tg256
119
tokens/s
30.1
tokens/s
34.33
sec
pp2048+tg256
141
tokens/s
34.7
tokens/s
14.52
sec
pp2048+tg768
142
tokens/s
36.0
tokens/s
14.40
sec
pp1024+tg1024
148
tokens/s
38.1
tokens/s
6.93
sec
pp1280+tg3072
151
tokens/s
33.2
tokens/s
8.50
sec
pp384+tg1152
159
tokens/s
42.1
tokens/s
2.43
sec
pp64+tg1024
141
tokens/s
43.4
tokens/s
472
ms
pp16+tg1536
144
tokens/s
43.7
tokens/s
130
ms