TEST #3112 RESULTS
03/18/2026 - 6:53 PM
ACCELERATOR
MODEL
180
tokens/s
generation
307
ms
time to first token
3827
tokens/s
prompt
1309
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 12P+4E
RAM
64GB
OS
Darwin
Kernel Release
25.3.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.3.0: Wed Jan 28 20:51:28 PST 2026; root:xnu-12377.91.3~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
5094
tokens/s
200
tokens/s
206
ms
pp4096+tg256
4213
tokens/s
123
tokens/s
980
ms
pp2048+tg256
4652
tokens/s
163
tokens/s
446
ms
pp2048+tg768
4644
tokens/s
156
tokens/s
447
ms
pp1024+tg1024
4679
tokens/s
182
tokens/s
224
ms
pp1280+tg3072
4432
tokens/s
148
tokens/s
294
ms
pp384+tg1152
4771
tokens/s
210
tokens/s
85
ms
pp64+tg1024
1423
tokens/s
224
tokens/s
49
ms
pp16+tg1536
539
tokens/s
215
tokens/s
34
ms