TEST #2997 RESULTS
03/07/2026 - 9:51 PM
ACCELERATOR
MODEL
118
tokens/s
generation
579
ms
time to first token
2068
tokens/s
prompt
751
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 10P+4E
RAM
48GB
OS
Darwin
Kernel Release
25.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.4.0: Wed Feb 25 21:02:10 PST 2026; root:xnu-12377.100.630.501.3~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2615
tokens/s
134
tokens/s
399
ms
pp4096+tg256
2149
tokens/s
77.5
tokens/s
1.92
sec
pp2048+tg256
2457
tokens/s
107
tokens/s
843
ms
pp2048+tg768
2451
tokens/s
102
tokens/s
845
ms
pp1024+tg1024
2545
tokens/s
121
tokens/s
410
ms
pp1280+tg3072
2525
tokens/s
95.6
tokens/s
515
ms
pp384+tg1152
2252
tokens/s
135
tokens/s
178
ms
pp64+tg1024
1125
tokens/s
151
tokens/s
63
ms
pp16+tg1536
492
tokens/s
144
tokens/s
38
ms