Related Models
Performance
331
tokens/sec
Faster than 97% of models
1.03
seconds
Faster than 54% of models
1.03
seconds
Faster than 68% of models
Market Median
94 tok/s
253% faster
Median TTFT
1.10s
6% faster
Speed Comparison
gpt-oss-120b (low)
327 tok/s-1%
Google: Gemini 3.1 Flash Lite Preview
324 tok/s-2%
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)
347 tok/s+5%
Benchmarks
MMLU-Pro
29.8%
GPQA Diamond
30.6%
HLE
5.2%
LiveCodeBench
8.1%
SciCode
2.5%
TerminalBench Hard
0.8%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025
8.3%
IFBench
19.5%
Long Context Recall
0.0%
Tau2
13.5%
Market AverageTop Score