TEST #1905 RESULTS
12/03/2025 - 8:43 PM
ACCELERATOR
MODEL
33.1
tokens/s
generation
3.38
sec
time to first token
366
tokens/s
prompt
153
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
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
418
tokens/s
35.9
tokens/s
2.48
sec
pp4096+tg256
380
tokens/s
25.4
tokens/s
10.81
sec
pp2048+tg256
406
tokens/s
31.4
tokens/s
5.08
sec
pp2048+tg768
405
tokens/s
30.5
tokens/s
5.09
sec
pp1024+tg1024
416
tokens/s
34.0
tokens/s
2.49
sec
pp1280+tg3072
412
tokens/s
29.1
tokens/s
3.13
sec
pp384+tg1152
411
tokens/s
36.4
tokens/s
960
ms
pp64+tg1024
329
tokens/s
38.2
tokens/s
219
ms
pp16+tg1536
116
tokens/s
37.2
tokens/s
162
ms