Skip to content
AI Atlas
Familyroot · LFM

LFM2.5

All 17 models in the terminal →

first seen 11 Sept 2026

family_01M2A55TZG2YQ10DPQYY7AB1H8

Models
17
First release
5 Jan 2026
Latest release
11 Aug 2026
Parameters
229.7M – 8.47B
Artifacts
10
Providers
1

Releases

Family releases by statusActiveFeb 26Mar 26Apr 26May 26Jun 26Jul 26Aug 26LFM2.5-VL-1.6B — 5 Jan 2026LFM2.5-1.2B-Instruct — 6 Jan 2026LFM2.5-350M — 31 Mar 2026LFM2.5-VL-450M — 8 Apr 2026LFM2.5-8B-A1B — 28 May 2026LFM2.5-230M — 24 Jun 2026LFM2.5-2.6B (free) — 28 Jul 2026LFM2.5-VL-3B — 11 Aug 2026

8 dated releases · dot size = best benchmark rank of the member · 9 members without a sourced release date not shown

Members 17

Family members
ModelParamsContextOpennessLicenceReleasedStatusBest rankCompare
LFM2.5-VL-3BLiquid AI3.12B—Open weightsOther11 Aug 2026Active—
LFM2.5-2.6B (free)Liquid AI2.7B128KOpen weightsOther28 Jul 2026Active#126 SciCode
LFM2.5-230MLiquid AI229.7M—Open weightsOther24 Jun 2026Active—
LFM2.5-8B-A1BLiquid AI8.47B1B active32.8KOpen weightsOther28 May 2026Active#120 IFBench
LFM2.5-VL-450MLiquid AI448.7M—Open weightsOther8 Apr 2026Active—
LFM2.5-350MLiquid AI354.5M—Open weightsOther31 Mar 2026Active—
LFM2.5-1.2B-InstructLiquid AI1.17B32KOpen weightsOther6 Jan 2026Active#174 IFBench
LFM2.5-VL-1.6BLiquid AI1.6B32KOpen weightsOther5 Jan 2026Active#152 MMMU-Pro
lfm2-5-1-2b-thinkingLiquid AI—32KOpen weights——Active#187 IFBench
LFM2.5-1.2B-BaseLiquid AI——Open weights——Active—
LFM2.5-2.6B-BaseLiquid AI——Open weights——Active—
LFM2.5-2.6B-DSparkLiquid AI——Open weights——Active—
LFM2.5-230M-BaseLiquid AI——Open weights——Active—
LFM2.5-350M-BaseLiquid AI——Open weights——Active—
LFM2.5-8B-A1B-BaseLiquid AI——Open weights——Active—
LFM2.5-8B-A1B-DSparkLiquid AI——Open weights——Active—
LFM2.5-Audio-1.5BLiquid AI——Open weights——Active—

Lineage among members 8 relations

Open in Graph →
Lineage of the LFM2.5 familyLFM2.5-VL-450M fine-tuned from LFM2.5-350MLFM2.5-VL-1.6B fine-tuned from LFM2.5-1.2B-BaseLFM2.5-1.2B-Instruct fine-tuned from LFM2.5-1.2B-BaseLFM2.5-2.6B (free) fine-tuned from LFM2.5-2.6B-BaseLFM2.5-VL-3B fine-tuned from LFM2.5-2.6B-BaseLFM2.5-350M fine-tuned from LFM2.5-350M-BaseLFM2.5-8B-A1B fine-tuned from LFM2.5-8B-A1B-BaseLFM2.5-230M fine-tuned from LFM2.5-230M-BaseLFM2.5-VL-1.6B1.6BLFM2.5-VL-1.6BLFM2.5-1.2B-Instruct1.17BLFM2.5-1.2B-InstructLFM2.5-350M354.5MLFM2.5-350MLFM2.5-VL-450M448.7MLFM2.5-VL-450MLFM2.5-8B-A1B8.47BLFM2.5-8B-A1BLFM2.5-230M229.7MLFM2.5-230MLFM2.5-2.6B (free)2.7BLFM2.5-2.6B (free)LFM2.5-VL-3B3.12BLFM2.5-VL-3BLFM2.5-1.2B-BaseLFM2.5-1.2B-BaseLFM2.5-2.6B-BaseLFM2.5-2.6B-BaseLFM2.5-230M-BaseLFM2.5-230M-BaseLFM2.5-350M-BaseLFM2.5-350M-BaseLFM2.5-8B-A1B-BaseLFM2.5-8B-A1B-Base

Benchmark progress

IFBenchbest #120

Best rank of the family on IFBench#300#200#100#0Feb 26Mar 26Apr 26May 26

Terminal-Benchbest #208

Best rank of the family on Terminal-Bench#400#300#200#100#0Feb 26Mar 26Apr 26May 26

Humanity's Last Exambest #244

Best rank of the family on Humanity's Last Exam#400#300#200#100#0Feb 26Mar 26Apr 26May 26Jun 26Jul 26

τ²-benchbest #266

Best rank of the family on τ²-bench#400#300#200#100#0Feb 26Mar 26Apr 26May 26

Artificial Analysis Intelligence Indexbest #278

Best rank of the family on Artificial Analysis Intelligence Index#600#400#200#0Feb 26Mar 26Apr 26May 26Jun 26Jul 26

GPQA Diamondbest #303

Best rank of the family on GPQA Diamond#500#400#300#200#100#0Feb 26Mar 26Apr 26May 26Jun 26Jul 26

Best rank (primary comparability group, current leaderboard) reached by any member released up to each date — lower is better, axis inverted. Ranks are today's ranks, not the ranks at release time.

Aggregates are computed from the members' sourced claims; members without a release date or parameter count are left out of the range and timeline rather than guessed. Methodology →