TEST #1615 RESULTS
10/07/2025 - 10:53 PM
ACCELERATOR
MODEL
8.4
tokens/s
generation
11.94
sec
time to first token
106
tokens/s
prompt
42
LocalScore
HOW YOU STACK UP
Explore All ResultsDeepSeek R1 Distill Qwen 1.5B - IQ4_XS - 4.25 bpw
SYSTEM
CPU
Intel Core i5-7400 CPU @ 3.00GHz (skylake)
RAM
15.6GB
OS
Linux
Kernel Release
6.8.0-85-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #85-Ubuntu SMP PREEMPT_DYNAMIC Thu Sep 18 15:26:59 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
113
tokens/s
8.6
tokens/s
9.19
sec
pp4096+tg256
112
tokens/s
7.9
tokens/s
36.68
sec
pp2048+tg256
113
tokens/s
8.3
tokens/s
18.23
sec
pp2048+tg768
113
tokens/s
8.2
tokens/s
18.23
sec
pp1024+tg1024
114
tokens/s
8.4
tokens/s
9.10
sec
pp1280+tg3072
113
tokens/s
8.2
tokens/s
11.43
sec
pp384+tg1152
114
tokens/s
8.5
tokens/s
3.49
sec
pp64+tg1024
99
tokens/s
8.7
tokens/s
759
ms
pp16+tg1536
68
tokens/s
8.6
tokens/s
349
ms