TEST #2254 RESULTS
01/12/2026 - 11:49 AM
ACCELERATOR
MODEL
258
tokens/s
generation
161
ms
time to first token
8757
tokens/s
prompt
2414
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
QEMU Virtual CPU version 2.5+
RAM
7.8GB
OS
Linux
Kernel Release
6.8.0-90-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #91-Ubuntu SMP PREEMPT_DYNAMIC Tue Nov 18 14:14:30 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
10774
tokens/s
273
tokens/s
99
ms
pp4096+tg256
6855
tokens/s
220
tokens/s
602
ms
pp2048+tg256
9082
tokens/s
251
tokens/s
230
ms
pp2048+tg768
9086
tokens/s
247
tokens/s
229
ms
pp1024+tg1024
10923
tokens/s
263
tokens/s
97
ms
pp1280+tg3072
10291
tokens/s
241
tokens/s
128
ms
pp384+tg1152
12197
tokens/s
272
tokens/s
35
ms
pp64+tg1024
8129
tokens/s
280
tokens/s
11
ms
pp16+tg1536
1476
tokens/s
277
tokens/s
14
ms