TEST #2228 RESULTS
01/10/2026 - 7:03 PM
ACCELERATOR
MODEL
20.2
tokens/s
generation
2.03
sec
time to first token
651
tokens/s
prompt
187
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7800X3D 8-Core Processor (znver4)
RAM
61.9GB
OS
Linux
Kernel Release
6.17.9-76061709-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #202511241048~1764607909~24.04~df6b2b6 SMP PREEMPT_DYNAMIC Mon D
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
708
tokens/s
23.8
tokens/s
1.49
sec
pp4096+tg256
617
tokens/s
10.8
tokens/s
6.73
sec
pp2048+tg256
696
tokens/s
19.5
tokens/s
2.99
sec
pp2048+tg768
695
tokens/s
18.0
tokens/s
3.00
sec
pp1024+tg1024
738
tokens/s
21.4
tokens/s
1.43
sec
pp1280+tg3072
719
tokens/s
15.0
tokens/s
1.83
sec
pp384+tg1152
759
tokens/s
23.7
tokens/s
545
ms
pp64+tg1024
627
tokens/s
25.4
tokens/s
138
ms
pp16+tg1536
303
tokens/s
24.5
tokens/s
88
ms