LiquidAI
LFM2.5 is a new family of hybrid models designed for on-device deployment. It builds on the LFM2 architecture with extended pre-training and reinforcement learning.
Running this yourself: can likely run on your own machine.
---
Quality Score
---
Arena ELO
3B
Parameters
128K
Context
This measures the amount of verifiable public evidence we have, not how capable the model is. A missing field means it has not been verified yet, not that its value is zero.
15 of 22 public signals
Sign in to join the discussion
27.8K
Downloads
49
Likes
Aug 2026
Released
5/5 signals
0/4 signals
4/5 signals
3/4 signals
3/4 signals
Parameters
3B
Training compute
Not reported
Dataset scale
Not reported
Base model
Not reported
Gaps we are still tracking
Metadata sources
Open Source
1
Recent launch, pricing, benchmark, and API signals linked to this model or its provider.
LFM2.5-2.6B-Base is now available through local Ollama runtime. 125K context window listed. LFM2.5-8B-A1B, an edge model built for fast, reliable tool calling on consumer hardware.
View sourceLFM2.5-2.6B-Base is now available through local Ollama runtime. 125K context window listed. LFM2.5-8B-A1B, an edge model built for fast, reliable tool calling on consumer hardware.