TEST #2588 RESULTS
02/10/2026 - 11:31 AM
ACCELERATOR
MODEL
242
tokens/s
generation
194
ms
time to first token
7283
tokens/s
prompt
2087
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 3950X 16-Core Processor (znver2)
RAM
94.2GB
OS
Linux
Kernel Release
6.8.0-94-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #96-Ubuntu SMP PREEMPT_DYNAMIC Fri Jan 9 20:36:55 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
8612
tokens/s
260
tokens/s
123
ms
pp4096+tg256
5700
tokens/s
201
tokens/s
723
ms
pp2048+tg256
7525
tokens/s
236
tokens/s
276
ms
pp2048+tg768
7490
tokens/s
231
tokens/s
278
ms
pp1024+tg1024
8799
tokens/s
246
tokens/s
120
ms
pp1280+tg3072
8308
tokens/s
223
tokens/s
158
ms
pp384+tg1152
10406
tokens/s
259
tokens/s
41
ms
pp64+tg1024
7512
tokens/s
266
tokens/s
12
ms
pp16+tg1536
1197
tokens/s
260
tokens/s
17
ms