# LFM2.5 230M

> Canonical: https://www.overmindlab.ai/models/lfm-2-5-230m

Tiny edge agent floor — 230M hybrid that still does tool use.

| Property | Value |
| --- | --- |
| Model ID | LiquidAI/LFM2.5-230M |
| Provider | Liquid AI |
| Group | LFM2.5 |
| Parameters | 230M |
| Tier | Compact |
| Context window | 32K |
| Max training context | 32K |
| Tool calling | Yes |
| Training methods | — |
| Training cost | from $0.50 |
| Serving cost (per 1M output tokens) | from $2.50 |

## Overview

LFM2.5 is Liquid AI's hybrid conv+attention family for edge and agentic workloads. The 230M tier is the smallest instruct checkpoint — distilled for tool use and data extraction under tight memory budgets.

Liquid AI hybrid models for on-device and edge deployment. Fast decode, low memory, and agentic post-training across the ladder.

## Good for

- On-device and edge tool-calling loops
- High-volume data extraction at minimal serve cost
- First experiment on the LFM2.5 ladder

Train this model on your agent's data: https://docs.overmindlab.ai/models/training.md

## Related models

- [LFM2.5 1.2B Instruct](https://www.overmindlab.ai/models/lfm-2-5-1-2b-instruct): 1.2B, 32K context
- [LFM2.5 350M](https://www.overmindlab.ai/models/lfm-2-5-350m): 350M, 32K context