TEST #2574 RESULTS
02/10/2026 - 9:04 AM
ACCELERATOR
MODEL
615
tokens/s
generation
25
ms
time to first token
59284
tokens/s
prompt
11290
LocalScore
HOW YOU STACK UP
Explore All ResultsPythia 70m Deduped - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-14450HX (alderlake)
RAM
15.7GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
70873
tokens/s
508
tokens/s
17
ms
pp4096+tg256
44890
tokens/s
586
tokens/s
94
ms
pp2048+tg256
73195
tokens/s
645
tokens/s
30
ms
pp2048+tg768
72022
tokens/s
609
tokens/s
31
ms
pp1024+tg1024
88368
tokens/s
639
tokens/s
13
ms
pp1280+tg3072
83338
tokens/s
614
tokens/s
17
ms
pp384+tg1152
80932
tokens/s
645
tokens/s
6
ms
pp64+tg1024
14875
tokens/s
658
tokens/s
17
ms
pp16+tg1536
5062
tokens/s
630
tokens/s
5
ms