TEST #2947 RESULTS
03/02/2026 - 10:39 AM
ACCELERATOR
MODEL
51.2
tokens/s
generation
2.04
sec
time to first token
590
tokens/s
prompt
246
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 12P+4E
RAM
128GB
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 20:48:41 PST 2026; root:xnu-12377.81.4~5/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
812
tokens/s
60.8
tokens/s
1.28
sec
pp4096+tg256
686
tokens/s
41.3
tokens/s
5.99
sec
pp2048+tg256
644
tokens/s
50.4
tokens/s
3.20
sec
pp2048+tg768
619
tokens/s
48.8
tokens/s
3.33
sec
pp1024+tg1024
658
tokens/s
52.9
tokens/s
1.57
sec
pp1280+tg3072
614
tokens/s
45.3
tokens/s
2.10
sec
pp384+tg1152
635
tokens/s
54.7
tokens/s
623
ms
pp64+tg1024
473
tokens/s
55.4
tokens/s
151
ms
pp16+tg1536
165
tokens/s
51.3
tokens/s
112
ms