TEST #2440 RESULTS
02/01/2026 - 5:12 AM
ACCELERATOR
MODEL
3091
tokens/s
generation
6
ms
time to first token
220409
tokens/s
prompt
47666
LocalScore
HOW YOU STACK UP
Explore All ResultsTiny Llama Ultra Compact - Q4_K - Medium
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
326385
tokens/s
3521
tokens/s
3
ms
pp4096+tg256
176934
tokens/s
2697
tokens/s
24
ms
pp2048+tg256
280356
tokens/s
2853
tokens/s
8
ms
pp2048+tg768
266445
tokens/s
2741
tokens/s
8
ms
pp1024+tg1024
319730
tokens/s
3059
tokens/s
3
ms
pp1280+tg3072
330229
tokens/s
2929
tokens/s
4
ms
pp384+tg1152
223750
tokens/s
3251
tokens/s
2
ms
pp64+tg1024
51220
tokens/s
3334
tokens/s
2
ms
pp16+tg1536
8634
tokens/s
3432
tokens/s
2
ms