LFM2.5 review, price and benchmarks (2026)

public free open-weights info-only local speed

Provider

Statuspublic

AA Intelligence Index8.3 (July 16, 2026 snapshot)

Releasedn/a

Context windown/a

Modalitiesn/a

API price $/1M (in / out)n/a

Speedn/a

AccessOpen weights · edge/on-device

In one line

'Liquid' non-Transformer architecture squeezing real capability into phone-sized models - watch this space for on-device AI.

Coding18

Reasoning16

Writing18

Agents & tools12

Category scores (0-100) are WhichAI blended ratings: public category leaderboards plus editorial judgment. Guidance, not official measurements.

Alternatives worth a look

Solar Pro 2 (Upstage) · AA 7.8API · Korean ecosystem
Llama 4 Scout (Meta) · AA 10Open weights · very long context
Ministral 3 14B (Mistral AI) · AA 11.1Open weights · edge/on-device

Open LFM2.5's official app ↗ Open in the WhichAI app Not sure? Take the 30-second finder

Artificial Analysis Intelligence Index, 2026 rebased scale (top models ≈ 55–60). Snapshot: July 24, 2026 (BenchLM mirror), retrieved July 26, 2026. Context, price and release data come from vendor pages and public pricing mirrors (retrieved July 19, 2026). Missing values are not published or not yet verified - shown as -. Sources are linked inside the app (Model guide and About > Methodology).