TEST #3183 RESULTS
03/25/2026 - 12:38 AM
ACCELERATOR
MODEL
44.9
tokens/s
generation
3.00
sec
time to first token
411
tokens/s
prompt
183
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M5 Pro 6P+12E
RAM
24GB
OS
Darwin
Kernel Release
25.3.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.3.0: Wed Jan 28 20:54:38 PST 2026; root:xnu-12377.91.3~2/RELEASE_ARM64_T6050
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
469
tokens/s
47.2
tokens/s
2.20
sec
pp4096+tg256
430
tokens/s
37.4
tokens/s
9.56
sec
pp2048+tg256
456
tokens/s
43.6
tokens/s
4.52
sec
pp2048+tg768
456
tokens/s
42.7
tokens/s
4.52
sec
pp1024+tg1024
466
tokens/s
45.8
tokens/s
2.22
sec
pp1280+tg3072
463
tokens/s
41.5
tokens/s
2.79
sec
pp384+tg1152
460
tokens/s
48.1
tokens/s
854
ms
pp64+tg1024
366
tokens/s
49.3
tokens/s
195
ms
pp16+tg1536
138
tokens/s
48.7
tokens/s
136
ms