TEST #2800 RESULTS
02/23/2026 - 4:14 AM
ACCELERATOR
MODEL
136
tokens/s
generation
147
ms
time to first token
9907
tokens/s
prompt
2090
LocalScore
HOW YOU STACK UP
Explore All Resultsh2o-danube3-500m-chat - F16
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
13380
tokens/s
141
tokens/s
84
ms
pp4096+tg256
7379
tokens/s
125
tokens/s
564
ms
pp2048+tg256
10296
tokens/s
136
tokens/s
206
ms
pp2048+tg768
10317
tokens/s
134
tokens/s
206
ms
pp1024+tg1024
13863
tokens/s
138
tokens/s
81
ms
pp1280+tg3072
12519
tokens/s
132
tokens/s
110
ms
pp384+tg1152
15805
tokens/s
139
tokens/s
31
ms
pp64+tg1024
4042
tokens/s
138
tokens/s
24
ms
pp16+tg1536
1558
tokens/s
138
tokens/s
18
ms