TEST #2920 RESULTS
02/27/2026 - 12:22 PM
ACCELERATOR
MODEL
32.8
tokens/s
generation
3.41
sec
time to first token
366
tokens/s
prompt
152
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 10P+4E
RAM
48GB
OS
Darwin
Kernel Release
25.1.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.1.0: Mon Oct 20 19:34:05 PDT 2025; root:xnu-12377.41.6~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
413
tokens/s
35.9
tokens/s
2.51
sec
pp4096+tg256
377
tokens/s
25.2
tokens/s
10.90
sec
pp2048+tg256
403
tokens/s
31.2
tokens/s
5.12
sec
pp2048+tg768
403
tokens/s
30.3
tokens/s
5.12
sec
pp1024+tg1024
413
tokens/s
33.6
tokens/s
2.51
sec
pp1280+tg3072
403
tokens/s
28.3
tokens/s
3.21
sec
pp384+tg1152
399
tokens/s
36.1
tokens/s
988
ms
pp64+tg1024
346
tokens/s
37.9
tokens/s
209
ms
pp16+tg1536
136
tokens/s
36.6
tokens/s
143
ms