TEST #3049 RESULTS
03/09/2026 - 1:31 AM
ACCELERATOR
MODEL
36.7
tokens/s
generation
1.17
sec
time to first token
1087
tokens/s
prompt
324
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2650 v2 @ 2.60GHz (ivybridge)
RAM
122.8GB
OS
Linux
Kernel Release
6.14.0-37-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #37-Ubuntu SMP PREEMPT_DYNAMIC Fri Nov 14 22:10:32 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1283
tokens/s
39.2
tokens/s
824
ms
pp4096+tg256
1066
tokens/s
31.1
tokens/s
3.87
sec
pp2048+tg256
1201
tokens/s
35.8
tokens/s
1.73
sec
pp2048+tg768
1206
tokens/s
35.2
tokens/s
1.73
sec
pp1024+tg1024
1277
tokens/s
37.5
tokens/s
828
ms
pp1280+tg3072
1235
tokens/s
33.6
tokens/s
1.06
sec
pp384+tg1152
1265
tokens/s
38.6
tokens/s
328
ms
pp64+tg1024
965
tokens/s
39.8
tokens/s
90
ms
pp16+tg1536
283
tokens/s
39.1
tokens/s
80
ms