lfm2.5-thinking:latest

Upstream reference: LFM2.5-1.2B-Thinking

Publisher: Liquid AI

Liquid AI describes this text-only reasoning checkpoint as part of the LFM2.5 hybrid-model family designed for on-device deployment.

Publisher-described context, separate from Sigma5C observations and recorded catalog statistics.

Published upstream specifications

Parameters

1.17 billion Source

Architecture blocks

10 double-gated convolution blocks and 6 grouped-query-attention blocks Source

Context

32,768 tokens Source

Knowledge cutoff stated by publisher

Mid-2024 Source

Upstream license label

lfm1.0 custom license; see the linked checkpoint license Source

Publisher-described uses

• General-purpose text reasoning described by Liquid AI. Source

Reference limitations

• Vision-language and audio-language LFM2.5 models are separate releases; their modalities do not apply to this text-only checkpoint. Source

A recorded tag and an upstream card association do not verify the deployed checkpoint, quantization, configured context or current hosted availability.

Publisher sources

• LFM2.5-1.2B-Thinking model card — Liquid AI. Checked 2026-11-02.

Links

Recorded references do not establish downloadable weights, licensing or current availability.

• Liquid AI — official site
• Ollama
• Hugging Face

Where lfm2.5-thinking:latest comes from

Named in 2 observation pages

• 3B-7B local models — all 53 · V1
• Models as data workers · V1