TEST #1643 RESULTS
10/13/2025 - 11:44 AM
ACCELERATOR
MODEL
98.7
tokens/s
generation
86
ms
time to first token
15044
tokens/s
prompt
2585
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
INTEL XEON PLATINUM 8568Y+ (sapphirerapids)
RAM
3023.6GB
OS
Linux
Kernel Release
6.8.0-60-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #63~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Apr 22 19:00:15 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
20155
tokens/s
102
tokens/s
61
ms
pp4096+tg256
16911
tokens/s
90.4
tokens/s
257
ms
pp2048+tg256
19207
tokens/s
96.8
tokens/s
117
ms
pp2048+tg768
19279
tokens/s
96.1
tokens/s
117
ms
pp1024+tg1024
20177
tokens/s
97.2
tokens/s
61
ms
pp1280+tg3072
17732
tokens/s
93.9
tokens/s
82
ms
pp384+tg1152
17053
tokens/s
101
tokens/s
31
ms
pp64+tg1024
3795
tokens/s
107
tokens/s
25
ms
pp16+tg1536
1085
tokens/s
104
tokens/s
23
ms