TEST #2034 RESULTS
12/23/2025 - 6:00 PM
ACCELERATOR
MODEL
88.7
tokens/s
generation
95
ms
time to first token
13774
tokens/s
prompt
2345
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon Gold 5418Y (sapphirerapids)
RAM
754.6GB
OS
Linux
Kernel Release
5.15.0-160-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #170-Ubuntu SMP Wed Oct 1 10:06:56 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
18679
tokens/s
91.9
tokens/s
66
ms
pp4096+tg256
15120
tokens/s
81.0
tokens/s
288
ms
pp2048+tg256
17737
tokens/s
87.7
tokens/s
127
ms
pp2048+tg768
17761
tokens/s
86.6
tokens/s
127
ms
pp1024+tg1024
18598
tokens/s
88.4
tokens/s
66
ms
pp1280+tg3072
16121
tokens/s
84.8
tokens/s
91
ms
pp384+tg1152
15511
tokens/s
89.5
tokens/s
35
ms
pp64+tg1024
3463
tokens/s
95.1
tokens/s
28
ms
pp16+tg1536
976
tokens/s
93.0
tokens/s
26
ms