Related Models
Performance
532
tokens/sec
Faster than 99% of models
1.25
seconds
Faster than 41% of models
1.25
seconds
Faster than 61% of models
Market Median
94 tok/s
464% faster
Median TTFT
1.11s
12% slower
Speed Comparison
LFM2.5-1.2B-Instruct
537 tok/s+1%
LFM2.5-VL-1.6B
508 tok/s-5%
Granite 4.0 H Small
441 tok/s-17%
Benchmarks
MMLU-Pro
25.7%
GPQA Diamond
22.8%
HLE
5.7%
LiveCodeBench
2.0%
SciCode
2.5%
TerminalBench Hard
0.0%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025
3.3%
IFBench
22.0%
Long Context Recall
0.0%
Tau2
12.6%
Market AverageTop Score