TEST #3057 RESULTS
03/11/2026 - 8:55 PM
ACCELERATOR
MODEL
46.8
tokens/s
generation
2.39
sec
time to first token
594
tokens/s
prompt
227
LocalScore
HOW YOU STACK UP
Explore All ResultsCodeLlama 7b Instruct Hf - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5800X3D 8-Core Processor (znver3)
RAM
63.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
736
tokens/s
60.2
tokens/s
1.41
sec
pp4096+tg256
444
tokens/s
9.5
tokens/s
9.33
sec
pp2048+tg256
627
tokens/s
48.1
tokens/s
3.28
sec
pp2048+tg768
622
tokens/s
45.1
tokens/s
3.31
sec
pp1024+tg1024
713
tokens/s
45.9
tokens/s
1.46
sec
pp1280+tg3072
701
tokens/s
31.4
tokens/s
1.84
sec
pp384+tg1152
732
tokens/s
57.8
tokens/s
540
ms
pp64+tg1024
278
tokens/s
63.9
tokens/s
244
ms
pp16+tg1536
495
tokens/s
59.4
tokens/s
46
ms