TEST #2047 RESULTS
12/26/2025 - 4:01 PM
ACCELERATOR
MODEL
28.1
tokens/s
generation
764
ms
time to first token
1780
tokens/s
prompt
403
LocalScore
HOW YOU STACK UP
Explore All ResultsEssentialAI_rnj 1 Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
119.7GB
OS
Linux
Kernel Release
6.14.0-1015-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #15-Ubuntu SMP PREEMPT_DYNAMIC Tue Nov 25 18:02:16 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2230
tokens/s
28.1
tokens/s
497
ms
pp4096+tg256
1567
tokens/s
26.9
tokens/s
2.65
sec
pp2048+tg256
1969
tokens/s
28.3
tokens/s
1.08
sec
pp2048+tg768
1948
tokens/s
27.4
tokens/s
1.09
sec
pp1024+tg1024
1927
tokens/s
27.4
tokens/s
577
ms
pp1280+tg3072
2154
tokens/s
27.4
tokens/s
635
ms
pp384+tg1152
2548
tokens/s
28.8
tokens/s
187
ms
pp64+tg1024
1282
tokens/s
29.0
tokens/s
85
ms
pp16+tg1536
393
tokens/s
29.2
tokens/s
75
ms