Gemma 3 1B

The Gemma 3 1B model is a lightweight, 1-billion-parameter language model by Google, optimized for efficiency on resource-limited devices. At 529MB, it processes text at 2,585 tokens/second with a context window of 128,000 tokens.

IFEval

80.2%

i
GSM8k

62.8%

i
Natural2Code

56.0%

i
MATH

48.0%

i
HumanEval

41.5%

i
BIG-Bench Hard

39.1%

i
FACTS Grounding

36.4%

i
WMT24++

35.9%

i
MBPP

35.2%

i
Global-MMLU-Lite

34.2%

i
GPQA

19.2%

i
HiddenMath

15.8%

i
MMLU-Pro

14.7%

i
BIG-Bench Extra Hard

7.2%

i
Bird-SQL (dev)

6.4%

i
SimpleQA

2.2%

i
LiveCodeBench

1.9%

i
ECLeKTic

1.4%

i