TEST #2822 RESULTS
02/24/2026 - 4:29 AM
ACCELERATOR
MODEL
204
tokens/s
generation
150
ms
time to first token
9703
tokens/s
prompt
2365
LocalScore
HOW YOU STACK UP
Explore All Resultsh2o-danube3-500m-chat - Q3_K - Small
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
11878
tokens/s
203
tokens/s
91
ms
pp4096+tg256
7063
tokens/s
187
tokens/s
585
ms
pp2048+tg256
9883
tokens/s
207
tokens/s
214
ms
pp2048+tg768
10135
tokens/s
206
tokens/s
207
ms
pp1024+tg1024
12877
tokens/s
208
tokens/s
84
ms
pp1280+tg3072
12408
tokens/s
196
tokens/s
108
ms
pp384+tg1152
16589
tokens/s
210
tokens/s
27
ms
pp64+tg1024
4072
tokens/s
213
tokens/s
20
ms
pp16+tg1536
2422
tokens/s
207
tokens/s
11
ms