TEST #3269 RESULTS
04/10/2026 - 3:48 AM
ACCELERATOR
MODEL
44.5
tokens/s
generation
3.01
sec
time to first token
410
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
64GB
OS
Darwin
Kernel Release
25.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.4.0: Thu Mar 19 19:33:50 PDT 2026; root:xnu-12377.101.15~1/RELEASE_ARM64_T6050
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
468
tokens/s
47.2
tokens/s
2.21
sec
pp4096+tg256
428
tokens/s
37.3
tokens/s
9.60
sec
pp2048+tg256
455
tokens/s
43.2
tokens/s
4.53
sec
pp2048+tg768
454
tokens/s
42.4
tokens/s
4.53
sec
pp1024+tg1024
468
tokens/s
45.4
tokens/s
2.21
sec
pp1280+tg3072
462
tokens/s
41.1
tokens/s
2.79
sec
pp384+tg1152
461
tokens/s
47.5
tokens/s
854
ms
pp64+tg1024
359
tokens/s
48.9
tokens/s
198
ms
pp16+tg1536
139
tokens/s
47.8
tokens/s
135
ms