Benchmark Almanac

LFM2.5-8B-A1B

LFM2.5-8B-A1B.

Liquid AI Released May 28th, 2026 Family LFM2.5 The lab's model card Compare with others
11
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Liquid AI May 2026

LFM2.5-8B-A1B

Open weights Mid-size

126.1 #309 of 404
Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.
  • IFBench 56.5% #59 /80
  • BFCL V4 48.5% #27 /30
  • LEXam-hard 15.5% #15 /15

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 91.8% #22 of 135 Lab model card
Multi-IF 79.9% #6 of 28 Lab model card
IFBench 56.5% #59 of 80 Lab model card
AA-Omniscience -24.7 #8 of 10 Lab model card

Math and science

BenchmarkScoreRankSource
MATH-500 88.8% #53 of 70 Lab model card
AIME 2025 42.5% #134 of 143 Lab model card
AIME 2026 50% #36 of 38 Lab model card

Agents and tools

BenchmarkScoreRankSource
Berkeley Function Calling Leaderboard (BFCL) 64.4% #31 of 41 Lab model card
tau2-bench 64% #64 of 79 Lab model card
BFCL V4 48.5% #27 of 30 Lab model card

Professional work

BenchmarkScoreRankSource
LEXam-hard 15.5% #15 of 15 Official leaderboard

Its lab's numbers put it at 126.1, number 309 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 123.0 to 129.1. How we compute it.

Its launch post

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