The user tested the newly released LiquidAI LFM2.5-24B-A2B model and reported strong performance, noting it builds on positive experiences with prior LiquidAI releases. They compared it to their regular Ornith 1.5-35B model, which they use in various workflows and as part of an AI‑Council in Pi Agent for its reliability and insight. After swapping LFM2.5-24B-A2B for Ornith in several Pi Agent tasks, they found it to be a capable replacement.