TEST #1738 RESULTS
11/05/2025 - 4:19 PM
ACCELERATOR
MODEL
32.7
tokens/s
generation
3.42
sec
time to first token
360
tokens/s
prompt
151
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 10P+4E
RAM
24GB
OS
Darwin
Kernel Release
24.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.6.0: Mon Jul 14 11:30:40 PDT 2025; root:xnu-11417.140.69~1/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
415
tokens/s
35.9
tokens/s
2.49
sec
pp4096+tg256
376
tokens/s
25.0
tokens/s
10.92
sec
pp2048+tg256
402
tokens/s
31.0
tokens/s
5.12
sec
pp2048+tg768
403
tokens/s
30.0
tokens/s
5.11
sec
pp1024+tg1024
414
tokens/s
33.4
tokens/s
2.50
sec
pp1280+tg3072
397
tokens/s
28.8
tokens/s
3.26
sec
pp384+tg1152
382
tokens/s
35.9
tokens/s
1.03
sec
pp64+tg1024
315
tokens/s
37.6
tokens/s
228
ms
pp16+tg1536
136
tokens/s
36.6
tokens/s
142
ms