LFM2-8B-A1B

LFM2-8B-A1B.

Liquid AI Released 7 Oct 2025 Family LFM2 The lab's model card

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Liquid AI Oct 2025

Open weights Mid-size

No Epoch Capabilities Index score yet

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Benchmark	Score	Rank	Source
MGSM	72.4%	#37 of 52	Model card
IFEval	77.6%	#105 of 126	Model card
MMLU	64.8%	#108 of 127	Model card
Multi-IF	58.2%	#24 of 28	Model card
Wer wird Millionär? LLM Benchmark	88.0	#69 of 76	Official leaderboard
MMMLU	55.3%	#57 of 62	Model card
IFBench	25.9%	#68 of 70	Model card
MMLU-Pro	37.4%	#147 of 151	Model card

Benchmark	Score	Rank	Source
GSM-Plus	64.8%	#1 of 2	Model card
MATH Level 5	62.4%	#1 of 2	Model card
GSM8K	84.4%	#53 of 87	Model card
MATH-500	74.2%	#61 of 67	Model card
GPQA Diamond	29.3%	#279 of 284	Model card

Benchmark	Score	Rank	Source
HumanEval+	69.5%	#23 of 27	Model card
LiveCodeBench v6	21%	#76 of 76	Model card
LiveCodeBench v5	21.4%	#44 of 44	Model card

Benchmark	Score		Rank	Source
SnakeBench	15		#165 of 183	Official leaderboard

The benchmarks the lab chose to headline, with the scores it gave.

Every result in the longer documents the lab published with it.