TEST #1481 RESULTS
09/05/2025 - 4:10 AM
ACCELERATOR
MODEL
25.8
tokens/s
generation
3.35
sec
time to first token
383
tokens/s
prompt
143
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M3 Max 12P+4E
RAM
64GB
OS
Darwin
Kernel Release
23.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 23.6.0: Thu Sep 12 23:36:23 PDT 2024; root:xnu-10063.141.1.701.1~1/RELEASE_ARM64_T6031
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
426
tokens/s
29.2
tokens/s
2.43
sec
pp4096+tg256
385
tokens/s
22.4
tokens/s
10.69
sec
pp2048+tg256
408
tokens/s
25.5
tokens/s
5.06
sec
pp2048+tg768
402
tokens/s
24.9
tokens/s
5.12
sec
pp1024+tg1024
419
tokens/s
26.0
tokens/s
2.47
sec
pp1280+tg3072
411
tokens/s
23.3
tokens/s
3.14
sec
pp384+tg1152
411
tokens/s
26.6
tokens/s
962
ms
pp64+tg1024
414
tokens/s
27.3
tokens/s
174
ms
pp16+tg1536
170
tokens/s
26.5
tokens/s
112
ms