TEST #3469 RESULTS
05/07/2026 - 11:28 PM
ACCELERATOR
13.6
tokens/s
generation
813
ms
time to first token
1537
tokens/s
prompt
295
LocalScore
HOW YOU STACK UP
Explore All ResultsMistral Nemo 2407 12B Thinking Claude Gemini GPT5.2 Uncensored HERETIC - Q6_K
SYSTEM
CPU
AMD Ryzen 9 7950X 16-Core Processor (znver4)
RAM
126.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2208
tokens/s
15.4
tokens/s
530
ms
pp4096+tg256
1751
tokens/s
10.7
tokens/s
2.42
sec
pp2048+tg256
1974
tokens/s
13.9
tokens/s
1.14
sec
pp2048+tg768
2194
tokens/s
14.6
tokens/s
1.00
sec
pp1024+tg1024
2246
tokens/s
15.1
tokens/s
519
ms
pp1280+tg3072
1949
tokens/s
10.2
tokens/s
736
ms
pp384+tg1152
778
tokens/s
13.6
tokens/s
627
ms
pp64+tg1024
558
tokens/s
15.2
tokens/s
187
ms
pp16+tg1536
172
tokens/s
13.5
tokens/s
163
ms