TEST #1476 RESULTS
09/05/2025 - 12:22 AM
ACCELERATOR
MODEL
188
tokens/s
generation
195
ms
time to first token
6899
tokens/s
prompt
1880
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 3700X 8-Core Processor (znver2)
RAM
31.9GB
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
8673
tokens/s
206
tokens/s
123
ms
pp4096+tg256
5808
tokens/s
136
tokens/s
712
ms
pp2048+tg256
7542
tokens/s
176
tokens/s
277
ms
pp2048+tg768
7466
tokens/s
182
tokens/s
280
ms
pp1024+tg1024
8711
tokens/s
199
tokens/s
123
ms
pp1280+tg3072
8290
tokens/s
178
tokens/s
159
ms
pp384+tg1152
8778
tokens/s
207
tokens/s
49
ms
pp64+tg1024
5557
tokens/s
206
tokens/s
17
ms
pp16+tg1536
1266
tokens/s
203
tokens/s
18
ms