TEST #782 RESULTS
04/29/2025 - 5:03 PM
ACCELERATOR
MODEL
68.3
tokens/s
generation
611
ms
time to first token
2034
tokens/s
prompt
610
LocalScore
HOW YOU STACK UP
Explore All ResultsDeepSeek R1 Distill Llama 8B - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 5600X 6-Core Processor (znver3)
RAM
95.9GB
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
2575
tokens/s
66.8
tokens/s
412
ms
pp4096+tg256
2112
tokens/s
58.4
tokens/s
1.96
sec
pp2048+tg256
2411
tokens/s
67.6
tokens/s
864
ms
pp2048+tg768
2441
tokens/s
65.9
tokens/s
855
ms
pp1024+tg1024
2528
tokens/s
70.7
tokens/s
420
ms
pp1280+tg3072
2561
tokens/s
66.0
tokens/s
514
ms
pp384+tg1152
2371
tokens/s
72.8
tokens/s
176
ms
pp64+tg1024
1238
tokens/s
73.5
tokens/s
65
ms
pp16+tg1536
71
tokens/s
73.1
tokens/s
238
ms