TEST #2287 RESULTS
01/15/2026 - 12:41 AM
ACCELERATOR
MODEL
308
tokens/s
generation
111
ms
time to first token
11982
tokens/s
prompt
3215
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon W-2133 CPU @ 3.60GHz (skylake-avx512)
RAM
125.4GB
OS
Linux
Kernel Release
6.14.0-37-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #37~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Nov 20 10:25:38 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
14898
tokens/s
323
tokens/s
72
ms
pp4096+tg256
10301
tokens/s
262
tokens/s
401
ms
pp2048+tg256
13340
tokens/s
300
tokens/s
157
ms
pp2048+tg768
13310
tokens/s
294
tokens/s
157
ms
pp1024+tg1024
15518
tokens/s
313
tokens/s
69
ms
pp1280+tg3072
14633
tokens/s
286
tokens/s
91
ms
pp384+tg1152
15927
tokens/s
324
tokens/s
27
ms
pp64+tg1024
8660
tokens/s
338
tokens/s
10
ms
pp16+tg1536
1254
tokens/s
332
tokens/s
16
ms