TEST #182 RESULTS
03/30/2025 - 9:06 PM
ACCELERATOR
MODEL
27.8
tokens/s
generation
4.33
sec
time to first token
293
tokens/s
prompt
123
LocalScore
HOW YOU STACK UP
Explore All ResultsDeepSeek R1 Distill Qwen 14B - Q4_K - Small
SYSTEM
CPU
Apple M4 Max 10P+4E
RAM
36GB
OS
Darwin
Kernel Release
24.1.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.1.0: Thu Nov 14 18:15:21 PST 2024; root:xnu-11215.41.3~13/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
321
tokens/s
29.8
tokens/s
3.23
sec
pp4096+tg256
288
tokens/s
21.5
tokens/s
14.27
sec
pp2048+tg256
321
tokens/s
26.6
tokens/s
6.41
sec
pp2048+tg768
320
tokens/s
25.8
tokens/s
6.44
sec
pp1024+tg1024
339
tokens/s
28.2
tokens/s
3.05
sec
pp1280+tg3072
322
tokens/s
24.7
tokens/s
4.02
sec
pp384+tg1152
346
tokens/s
30.4
tokens/s
1.14
sec
pp64+tg1024
283
tokens/s
31.8
tokens/s
255
ms
pp16+tg1536
99
tokens/s
31.0
tokens/s
191
ms
