TEST #1265 RESULTS
07/28/2025 - 9:15 PM
ACCELERATOR
MODEL
24.5
tokens/s
generation
4.20
sec
time to first token
288
tokens/s
prompt
119
LocalScore
HOW YOU STACK UP
Explore All ResultsDeepseek-R1-0528-Qwen3-8B - Q6_K
SYSTEM
CPU
Apple M1 Max 8P+2E
RAM
64GB
OS
Darwin
Kernel Release
24.5.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.5.0: Tue Apr 22 19:54:49 PDT 2025; root:xnu-11417.121.6~2/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
334
tokens/s
26.1
tokens/s
3.10
sec
pp4096+tg256
312
tokens/s
19.4
tokens/s
13.17
sec
pp2048+tg256
327
tokens/s
23.7
tokens/s
6.31
sec
pp2048+tg768
324
tokens/s
23.0
tokens/s
6.36
sec
pp1024+tg1024
331
tokens/s
25.3
tokens/s
3.13
sec
pp1280+tg3072
330
tokens/s
21.7
tokens/s
3.92
sec
pp384+tg1152
320
tokens/s
26.5
tokens/s
1.23
sec
pp64+tg1024
239
tokens/s
27.9
tokens/s
301
ms
pp16+tg1536
74
tokens/s
27.3
tokens/s
251
ms