TEST #98 RESULTS
03/28/2025 - 9:33 PM
ACCELERATOR
MODEL
97.0
tokens/s
generation
543
ms
time to first token
2378
tokens/s
prompt
752
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2620 0 @ 2.00GHz (sandybridge)
RAM
503.8GB
OS
Linux
Kernel Release
6.5.0-41-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #41~22.04.2-Ubuntu SMP PREEMPT_DYNAMIC Mon Jun 3 11:32:55 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2867
tokens/s
101
tokens/s
367
ms
pp4096+tg256
2215
tokens/s
67.4
tokens/s
1.86
sec
pp2048+tg256
2621
tokens/s
89.8
tokens/s
793
ms
pp2048+tg768
2609
tokens/s
86.3
tokens/s
796
ms
pp1024+tg1024
2867
tokens/s
100
tokens/s
367
ms
pp1280+tg3072
2712
tokens/s
80.9
tokens/s
482
ms
pp384+tg1152
2904
tokens/s
111
tokens/s
141
ms
pp64+tg1024
1986
tokens/s
121
tokens/s
40
ms
pp16+tg1536
624
tokens/s
116
tokens/s
33
ms