TEST #979 RESULTS
06/03/2025 - 2:46 AM
ACCELERATOR
MODEL
98.0
tokens/s
generation
423
ms
time to first token
3355
tokens/s
prompt
920
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - 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
4177
tokens/s
92.4
tokens/s
258
ms
pp4096+tg256
2609
tokens/s
89.0
tokens/s
1.58
sec
pp2048+tg256
3486
tokens/s
99.0
tokens/s
597
ms
pp2048+tg768
3479
tokens/s
96.2
tokens/s
599
ms
pp1024+tg1024
4137
tokens/s
102
tokens/s
257
ms
pp1280+tg3072
3913
tokens/s
94.3
tokens/s
336
ms
pp384+tg1152
4830
tokens/s
101
tokens/s
98
ms
pp64+tg1024
3074
tokens/s
103
tokens/s
34
ms
pp16+tg1536
487
tokens/s
104
tokens/s
44
ms