TEST #1698 RESULTS
10/25/2025 - 1:57 AM
ACCELERATOR
MODEL
16.6
tokens/s
generation
8.11
sec
time to first token
152
tokens/s
prompt
68
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 4P+6E
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 Aug 11 21:16:30 PDT 2025; root:xnu-11417.140.69.701.11~1/RELEASE_ARM64_T8132
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
184
tokens/s
18.9
tokens/s
5.63
sec
pp4096+tg256
163
tokens/s
15.0
tokens/s
25.16
sec
pp2048+tg256
169
tokens/s
17.3
tokens/s
12.15
sec
pp2048+tg768
162
tokens/s
16.4
tokens/s
12.69
sec
pp1024+tg1024
169
tokens/s
17.1
tokens/s
6.12
sec
pp1280+tg3072
164
tokens/s
15.0
tokens/s
7.84
sec
pp384+tg1152
151
tokens/s
16.5
tokens/s
2.60
sec
pp64+tg1024
141
tokens/s
16.7
tokens/s
510
ms
pp16+tg1536
61
tokens/s
16.1
tokens/s
320
ms