TEST #3001 RESULTS
03/07/2026 - 10:01 PM
ACCELERATOR
MODEL
1529
tokens/s
generation
11
ms
time to first token
124862
tokens/s
prompt
25867
LocalScore
HOW YOU STACK UP
Explore All Resultsllama-39m - 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
124768
tokens/s
1279
tokens/s
9
ms
pp4096+tg256
103346
tokens/s
1188
tokens/s
41
ms
pp2048+tg256
158697
tokens/s
1537
tokens/s
14
ms
pp2048+tg768
169927
tokens/s
1526
tokens/s
13
ms
pp1024+tg1024
178612
tokens/s
1715
tokens/s
6
ms
pp1280+tg3072
186055
tokens/s
1475
tokens/s
7
ms
pp384+tg1152
159913
tokens/s
1696
tokens/s
4
ms
pp64+tg1024
35858
tokens/s
1664
tokens/s
3
ms
pp16+tg1536
6578
tokens/s
1678
tokens/s
3
ms