TEST #2378 RESULTS
01/30/2026 - 12:25 PM
ACCELERATOR
MODEL
107
tokens/s
generation
281
ms
time to first token
5212
tokens/s
prompt
1257
LocalScore
HOW YOU STACK UP
Explore All ResultsDeepseek Coder 1.3b Instruct - 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
6468
tokens/s
115
tokens/s
167
ms
pp4096+tg256
3982
tokens/s
82.6
tokens/s
1.04
sec
pp2048+tg256
5103
tokens/s
102
tokens/s
411
ms
pp2048+tg768
5027
tokens/s
98.7
tokens/s
418
ms
pp1024+tg1024
6320
tokens/s
109
tokens/s
171
ms
pp1280+tg3072
5750
tokens/s
94.9
tokens/s
231
ms
pp384+tg1152
8183
tokens/s
119
tokens/s
55
ms
pp64+tg1024
4716
tokens/s
123
tokens/s
21
ms
pp16+tg1536
1358
tokens/s
120
tokens/s
19
ms