Liquid AILFM2.5
LFM2.5-1.2B-Thinking
LFM2.5-1.2B-Thinking is the reasoning variant of Liquid's 1.2B text model.
What the release provides
It uses a 32,768-token context and the LFM2.5 hybrid architecture. The separately named Instruct checkpoint serves a different response style.LFM2.5-1.2B-Thinking official model card
How to use it well
Practical guidance. Compare the reasoning variant on questions where the additional generation produces a measurable improvement. Record the final answer and the total latency. For short extraction tasks, check whether reasoning changes accuracy enough to justify the extra output.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | LiquidAI/LFM2.5-1.2B-Thinking |
| Context window | 32,768 tokens Downloadable checkpoint limit; the hosted API has a separate service contract. LFM2.5-1.2B-Thinking official model card |