TEST #2773 RESULTS
02/21/2026 - 7:43 PM
ACCELERATOR
MODEL
41.9
tokens/s
generation
488
ms
time to first token
2617
tokens/s
prompt
608
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 9950X3D 16-Core Processor
RAM
60.3GB
OS
Linux
Kernel Release
6.17.0-14-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #14~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Jan 15 15:52:10 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2410
tokens/s
43.1
tokens/s
447
ms
pp4096+tg256
2479
tokens/s
28.1
tokens/s
1.69
sec
pp2048+tg256
3373
tokens/s
37.6
tokens/s
634
ms
pp2048+tg768
3330
tokens/s
37.0
tokens/s
642
ms
pp1024+tg1024
3749
tokens/s
45.3
tokens/s
295
ms
pp1280+tg3072
3227
tokens/s
34.2
tokens/s
419
ms
pp384+tg1152
3675
tokens/s
49.1
tokens/s
124
ms
pp64+tg1024
804
tokens/s
52.1
tokens/s
97
ms
pp16+tg1536
507
tokens/s
50.5
tokens/s
49
ms