TEST #2621 RESULTS
02/14/2026 - 12:16 PM
ACCELERATOR
MODEL
195
tokens/s
generation
178
ms
time to first token
7724
tokens/s
prompt
2040
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5800X 8-Core Processor (znver3)
RAM
63.9GB
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
9691
tokens/s
198
tokens/s
111
ms
pp4096+tg256
6307
tokens/s
188
tokens/s
655
ms
pp2048+tg256
8313
tokens/s
201
tokens/s
252
ms
pp2048+tg768
8304
tokens/s
202
tokens/s
252
ms
pp1024+tg1024
9708
tokens/s
199
tokens/s
111
ms
pp1280+tg3072
9187
tokens/s
195
tokens/s
144
ms
pp384+tg1152
10703
tokens/s
196
tokens/s
41
ms
pp64+tg1024
5866
tokens/s
189
tokens/s
16
ms
pp16+tg1536
1436
tokens/s
191
tokens/s
16
ms