TEST #3398 RESULTS
04/27/2026 - 4:43 AM
ACCELERATOR
MODEL
90.6
tokens/s
generation
480
ms
time to first token
3043
tokens/s
prompt
831
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core 5 210H (alderlake)
RAM
15.6GB
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
3777
tokens/s
97.6
tokens/s
281
ms
pp4096+tg256
2258
tokens/s
80.2
tokens/s
1.83
sec
pp2048+tg256
3050
tokens/s
87.9
tokens/s
684
ms
pp2048+tg768
3050
tokens/s
85.4
tokens/s
683
ms
pp1024+tg1024
3623
tokens/s
91.7
tokens/s
294
ms
pp1280+tg3072
3406
tokens/s
84.6
tokens/s
387
ms
pp384+tg1152
4249
tokens/s
96.4
tokens/s
101
ms
pp64+tg1024
3309
tokens/s
97.1
tokens/s
29
ms
pp16+tg1536
670
tokens/s
94.8
tokens/s
33
ms