TEST #2816 RESULTS
02/24/2026 - 4:24 AM
ACCELERATOR
MODEL
28.4
tokens/s
generation
3.71
sec
time to first token
329
tokens/s
prompt
136
LocalScore
HOW YOU STACK UP
Explore All Resultsh2o-danube3-500m-chat - IQ4_XS - 4.25 bpw
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
392
tokens/s
28.9
tokens/s
2.65
sec
pp4096+tg256
344
tokens/s
26.7
tokens/s
12.02
sec
pp2048+tg256
385
tokens/s
27.2
tokens/s
5.43
sec
pp2048+tg768
377
tokens/s
28.1
tokens/s
5.53
sec
pp1024+tg1024
404
tokens/s
29.2
tokens/s
2.60
sec
pp1280+tg3072
366
tokens/s
27.8
tokens/s
3.59
sec
pp384+tg1152
386
tokens/s
29.5
tokens/s
1.03
sec
pp64+tg1024
198
tokens/s
29.0
tokens/s
358
ms
pp16+tg1536
113
tokens/s
29.3
tokens/s
183
ms