LFM2.5-1.2B-Instruct

Liquid AI

Verified against Artificial Analysis · Jul 27, 2026

Capability

percentile index

Price /M

$0 / $0

in / out

Context

—

Avg score

20.3

8 benchmarks

AA Intelligence

2.7

index

Weights

Closed

Released

Jan 5, 2026

About

LFM2.5-1.2B-Instruct is a proprietary model in the LFM2.5 family from Liquid AI. Benchmarked on 8 evals, averaging 20.3.

Benchmark scores · 8

avg 20.3 — higher is better

MATHartificial-analysis · 2026-07

71.7

IFBenchartificial-analysis · 2026-07

GPQA Diamondartificial-analysis · 2026-07

29.5

τ²-benchartificial-analysis · 2026-07

10.8

Humanity's Last Examartificial-analysis · 2026-07

6.8

SciCodeartificial-analysis · 2026-07

2.3

Terminal-Bench Hardartificial-analysis · 2026-07

AA Long-Context Reasoningartificial-analysis · 2026-07

Details

PublisherLiquid AI

FamilyLFM2.5

Parameters—

License—

WeightsClosed

ReleasedJan 5, 2026

History

full ledger →

Capability over time

GPQA Diamond32.6 → 29.5

yesterday

MATHfirst seen at 71.7

yesterday

IFBench43.8 → 41

yesterday

Append-only ledger — every observed change to this model's numbers.

API providers

Artificial Analysis$0/M