TEST #3111 RESULTS
03/18/2026 - 6:48 PM
ACCELERATOR
MODEL
41.8
tokens/s
generation
2.44
sec
time to first token
478
tokens/s
prompt
202
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 12P+4E
RAM
64GB
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 20:51:28 PST 2026; root:xnu-12377.91.3~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
809
tokens/s
60.6
tokens/s
1.28
sec
pp4096+tg256
677
tokens/s
41.0
tokens/s
6.07
sec
pp2048+tg256
607
tokens/s
50.2
tokens/s
3.40
sec
pp2048+tg768
539
tokens/s
40.1
tokens/s
3.82
sec
pp1024+tg1024
317
tokens/s
40.8
tokens/s
3.27
sec
pp1280+tg3072
472
tokens/s
26.7
tokens/s
2.74
sec
pp384+tg1152
399
tokens/s
35.2
tokens/s
992
ms
pp64+tg1024
353
tokens/s
39.7
tokens/s
202
ms
pp16+tg1536
130
tokens/s
41.6
tokens/s
143
ms