TEST #985 RESULTS
06/03/2025 - 3:58 AM
ACCELERATOR
MODEL
27.3
tokens/s
generation
1.42
sec
time to first token
911
tokens/s
prompt
260
LocalScore
HOW YOU STACK UP
Explore All ResultsDeepSeek R1 Distill Qwen 7B - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-4590 CPU @ 3.30GHz (haswell)
RAM
31.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1059
tokens/s
27.7
tokens/s
1.00
sec
pp4096+tg256
885
tokens/s
25.5
tokens/s
4.67
sec
pp2048+tg256
990
tokens/s
27.2
tokens/s
2.10
sec
pp2048+tg768
990
tokens/s
27.0
tokens/s
2.11
sec
pp1024+tg1024
1045
tokens/s
27.8
tokens/s
1.02
sec
pp1280+tg3072
1021
tokens/s
26.6
tokens/s
1.29
sec
pp384+tg1152
1113
tokens/s
27.9
tokens/s
380
ms
pp64+tg1024
880
tokens/s
28.1
tokens/s
127
ms
pp16+tg1536
219
tokens/s
28.0
tokens/s
107
ms