TEST #2557 RESULTS
02/07/2026 - 5:11 AM
ACCELERATOR
MODEL
39.4
tokens/s
generation
5.81
sec
time to first token
239
tokens/s
prompt
117
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
11th Gen Intel Core i9-11900H @ 2.50GHz (tigerlake)
RAM
31.1GB
OS
Linux
Kernel Release
6.8.0-100-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #100-Ubuntu SMP PREEMPT_DYNAMIC Tue Jan 13 16:40:06 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
310
tokens/s
43.5
tokens/s
3.32
sec
pp4096+tg256
222
tokens/s
34.5
tokens/s
18.52
sec
pp2048+tg256
237
tokens/s
38.8
tokens/s
8.65
sec
pp2048+tg768
226
tokens/s
38.1
tokens/s
9.10
sec
pp1024+tg1024
223
tokens/s
39.7
tokens/s
4.61
sec
pp1280+tg3072
214
tokens/s
36.7
tokens/s
6.00
sec
pp384+tg1152
225
tokens/s
40.9
tokens/s
1.73
sec
pp64+tg1024
241
tokens/s
42.0
tokens/s
289
ms
pp16+tg1536
250
tokens/s
40.5
tokens/s
86
ms