TEST #259 RESULTS
04/05/2025 - 12:40 AM
ACCELERATOR
MODEL
12.2
tokens/s
generation
21.42
sec
time to first token
64
tokens/s
prompt
33
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 Coder 3B Instruct - Q8_0
SYSTEM
CPU
Intel Xeon CPU E5-2697 v2 @ 2.70GHz (ivybridge)
RAM
251.8GB
OS
Linux
Kernel Release
6.11.0-14-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #15-Ubuntu SMP PREEMPT_DYNAMIC Fri Jan 10 23:48:25 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
63
tokens/s
12.4
tokens/s
16.43
sec
pp4096+tg256
60
tokens/s
11.5
tokens/s
68.79
sec
pp2048+tg256
63
tokens/s
12.0
tokens/s
32.44
sec
pp2048+tg768
63
tokens/s
12.1
tokens/s
32.58
sec
pp1024+tg1024
66
tokens/s
12.2
tokens/s
15.68
sec
pp1280+tg3072
65
tokens/s
11.9
tokens/s
19.67
sec
pp384+tg1152
67
tokens/s
12.4
tokens/s
5.85
sec
pp64+tg1024
67
tokens/s
12.5
tokens/s
1.03
sec
pp16+tg1536
59
tokens/s
12.5
tokens/s
349
ms