TEST #550 RESULTS
04/08/2025 - 1:43 AM
ACCELERATOR
MODEL
30.2
tokens/s
generation
4.14
sec
time to first token
302
tokens/s
prompt
130
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 8P+4E
RAM
48GB
OS
Darwin
Kernel Release
24.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.4.0: Wed Mar 19 21:17:35 PDT 2025; root:xnu-11417.101.15~1/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
337
tokens/s
32.8
tokens/s
3.07
sec
pp4096+tg256
312
tokens/s
23.7
tokens/s
13.18
sec
pp2048+tg256
331
tokens/s
28.9
tokens/s
6.23
sec
pp2048+tg768
329
tokens/s
28.0
tokens/s
6.25
sec
pp1024+tg1024
334
tokens/s
30.9
tokens/s
3.09
sec
pp1280+tg3072
333
tokens/s
26.8
tokens/s
3.87
sec
pp384+tg1152
330
tokens/s
32.9
tokens/s
1.19
sec
pp64+tg1024
294
tokens/s
34.3
tokens/s
245
ms
pp16+tg1536
121
tokens/s
33.3
tokens/s
159
ms