TEST #3990 RESULTS
08/03/2026 - 1:44 PM
ACCELERATOR
MODEL
34.9
tokens/s
generation
3.72
sec
time to first token
369
tokens/s
prompt
151
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2697 v3 @ 2.60GHz (haswell)
RAM
251.8GB
OS
Linux
Kernel Release
7.0.14-5-pve
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC PMX 7.0.14-5 (2026-07-14T12:32Z)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
248
tokens/s
40.0
tokens/s
4.15
sec
pp4096+tg256
313
tokens/s
31.2
tokens/s
13.13
sec
pp2048+tg256
377
tokens/s
33.9
tokens/s
5.45
sec
pp2048+tg768
453
tokens/s
33.5
tokens/s
4.55
sec
pp1024+tg1024
474
tokens/s
34.6
tokens/s
2.19
sec
pp1280+tg3072
455
tokens/s
33.2
tokens/s
2.84
sec
pp384+tg1152
443
tokens/s
35.8
tokens/s
894
ms
pp64+tg1024
340
tokens/s
36.2
tokens/s
216
ms
pp16+tg1536
216
tokens/s
35.5
tokens/s
101
ms