Benchmark Almanac

LFM2.5-1.2B-Thinking

LFM2.5-1.2B-Thinking.

Liquid AI Released January 20th, 2026 Family LFM2.5 The lab's model card Compare with others
9
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Liquid AI Jan 2026

LFM2.5-1.2B-Thinking

Open weights Small

119.0 #351 of 404
Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.
  • IFBench 44.9% #68 /80
  • MMLU-Pro 49.6% #155 /167
  • AIME 2025 31.7% #138 /143

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.

Knowledge and language

BenchmarkScoreRankSource
IFEval 88.4% #51 of 135 Lab model card
Multi-IF 69.3% #17 of 28 Lab model card
IFBench 44.9% #68 of 80 Lab model card
MMLU-Pro 49.6% #155 of 167 Lab model card

Math and science

BenchmarkScoreRankSource
GSM8K 85.6% #56 of 92 Lab model card
MATH-500 88% #54 of 70 Lab model card
GPQA Diamond 37.9% #278 of 302 Lab model card
AIME 2025 31.7% #138 of 143 Lab model card

Agents and tools

BenchmarkScoreRankSource
Berkeley Function Calling Leaderboard (BFCL) 57% #38 of 41 Lab model card

Its lab's numbers put it at 119.0, number 351 of 404

The Self-Reported Index fits one score to the 9 capability results Liquid AI reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 116.4 to 121.7. How we compute it.

Its launch post

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