TEST #2824 RESULTS
02/24/2026 - 4:31 AM
ACCELERATOR
MODEL
206
tokens/s
generation
145
ms
time to first token
10482
tokens/s
prompt
2459
LocalScore
HOW YOU STACK UP
Explore All Resultsh2o-danube3-500m-chat - Q4_1
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
12271
tokens/s
212
tokens/s
88
ms
pp4096+tg256
7373
tokens/s
186
tokens/s
562
ms
pp2048+tg256
9977
tokens/s
205
tokens/s
211
ms
pp2048+tg768
10267
tokens/s
207
tokens/s
204
ms
pp1024+tg1024
14035
tokens/s
210
tokens/s
77
ms
pp1280+tg3072
12297
tokens/s
203
tokens/s
109
ms
pp384+tg1152
17250
tokens/s
212
tokens/s
28
ms
pp64+tg1024
8461
tokens/s
211
tokens/s
12
ms
pp16+tg1536
2411
tokens/s
206
tokens/s
12
ms