TEST #26 RESULTS
03/27/2025 - 11:39 PM
ACCELERATOR
MODEL
310
tokens/s
generation
110
ms
time to first token
12253
tokens/s
prompt
3257
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7H12 64-Core Processor (znver2)
RAM
503.5GB
OS
Linux
Kernel Release
6.8.0-52-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #53~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Wed Jan 15 19:18:46 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
15745
tokens/s
323
tokens/s
68
ms
pp4096+tg256
10626
tokens/s
266
tokens/s
389
ms
pp2048+tg256
13305
tokens/s
303
tokens/s
158
ms
pp2048+tg768
13497
tokens/s
300
tokens/s
155
ms
pp1024+tg1024
15548
tokens/s
317
tokens/s
69
ms
pp1280+tg3072
14797
tokens/s
288
tokens/s
90
ms
pp384+tg1152
16781
tokens/s
326
tokens/s
26
ms
pp64+tg1024
9245
tokens/s
337
tokens/s
10
ms
pp16+tg1536
737
tokens/s
330
tokens/s
25
ms