TEST #2565 RESULTS
02/10/2026 - 8:53 AM
ACCELERATOR
MODEL
286
tokens/s
generation
155
ms
time to first token
11023
tokens/s
prompt
2730
LocalScore
HOW YOU STACK UP
Explore All ResultsTinyMistral 248M v2.5 - Q6_K
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
12200
tokens/s
268
tokens/s
87
ms
pp4096+tg256
6323
tokens/s
251
tokens/s
652
ms
pp2048+tg256
9814
tokens/s
281
tokens/s
213
ms
pp2048+tg768
9680
tokens/s
282
tokens/s
215
ms
pp1024+tg1024
13418
tokens/s
299
tokens/s
80
ms
pp1280+tg3072
12122
tokens/s
271
tokens/s
109
ms
pp384+tg1152
22207
tokens/s
305
tokens/s
20
ms
pp64+tg1024
10224
tokens/s
308
tokens/s
9
ms
pp16+tg1536
3222
tokens/s
309
tokens/s
8
ms