TEST #2310 RESULTS
01/16/2026 - 3:03 AM
ACCELERATOR
MODEL
59.5
tokens/s
generation
1.64
sec
time to first token
703
tokens/s
prompt
294
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 Ultra 16P+8E
RAM
64GB
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 11:30:30 PDT 2025; root:xnu-11417.140.69~1/RELEASE_ARM64_T6020
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
847
tokens/s
64.2
tokens/s
1.22
sec
pp4096+tg256
817
tokens/s
46.4
tokens/s
5.04
sec
pp2048+tg256
843
tokens/s
56.6
tokens/s
2.45
sec
pp2048+tg768
843
tokens/s
55.1
tokens/s
2.45
sec
pp1024+tg1024
848
tokens/s
60.6
tokens/s
1.22
sec
pp1280+tg3072
837
tokens/s
52.9
tokens/s
1.55
sec
pp384+tg1152
776
tokens/s
65.0
tokens/s
509
ms
pp64+tg1024
403
tokens/s
68.1
tokens/s
173
ms
pp16+tg1536
116
tokens/s
66.4
tokens/s
151
ms