TEST #53 RESULTS
03/28/2025 - 8:43 PM
ACCELERATOR
MODEL
313
tokens/s
generation
110
ms
time to first token
11835
tokens/s
prompt
3229
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7B13 64-Core Processor (znver3)
RAM
251.5GB
OS
Linux
Kernel Release
6.8.0-48-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #48~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Mon Oct 7 11:24:13 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
15256
tokens/s
319
tokens/s
70
ms
pp4096+tg256
10492
tokens/s
266
tokens/s
394
ms
pp2048+tg256
13463
tokens/s
306
tokens/s
156
ms
pp2048+tg768
13513
tokens/s
302
tokens/s
155
ms
pp1024+tg1024
15944
tokens/s
322
tokens/s
67
ms
pp1280+tg3072
14635
tokens/s
293
tokens/s
91
ms
pp384+tg1152
17270
tokens/s
330
tokens/s
25
ms
pp64+tg1024
4828
tokens/s
343
tokens/s
16
ms
pp16+tg1536
1113
tokens/s
338
tokens/s
17
ms