---
title: "Public sector | Solutions | Liquid AI"
description: "On-device AI for the public sector and defense: navigation, imagery, voice and onboard control that keep working without a network."
canonical: "https://www.liquid.ai/public-sector"
locale: "en"
---

# AI built for the real-world constraints of the public sector.

Bring tailored AI to agency systems and operators in the field. Build around your data, privacy requirements, and the conditions in which your teams operate.

[Contact sales](https://www.liquid.ai/connect?topic=enterprise)

Specialized for your mission, built for the environments your teams work in.

[Meet the models behind it](https://www.liquid.ai/models)

## Why Liquid for the public sector.

- [Private AI, deployed on your terms. — Bring AI to sovereign infrastructure or private cloud. Run and customize models within your data boundaries, with access governed by your systems. — Discuss your deployment](https://www.liquid.ai/connect?topic=enterprise)
- [Built for limited compute and connectivity. — Bring text, image, and audio capabilities to supported laptops, field devices, and embedded systems. Run compact models on the compute you already own, with local inference where connectivity is limited. — See the DPhi Space collaboration](https://payloadspace.com/dphis-second-demo-brings-compute-space-to-orbit/)
- [Specialized intelligence, built around your data. — Fine-tune models on your own documents and terminology to target more accurate outputs, from case summaries to field reports. — Explore the model library](https://www.liquid.ai/models)

## Example use cases, from agency systems to the field.

### Turn complex requests into cases ready for review.

Adapt Liquid models to gather information, identify missing details, and prepare case summaries. Connect them to agency procedures and systems, with staff reviewing exceptions and approving decisions.

Runs on sovereign infrastructure or private cloud, so citizen records stay inside government systems while claims are prepared.[Discuss your deployment](https://www.liquid.ai/connect?topic=enterprise)Illustrative example: The model reads a new unemployment claim, finds two employers, and flags conflicting dates and a missing document for the caseworker. Fictional data; actual workflows depend on your systems and integrations.

## Further reading

- [ — OCT 27, 2026Models — Introducing LFM3: Every Modality for Specialized Intelligence](https://www.liquid.ai/blog/lfm3)
- [ — MAR 5, 2026Models — No Cloud, No Waiting: Tool-Calling Agents on Consumer Hardware with LFM2-24B-A2B](https://www.liquid.ai/blog/no-cloud-tool-calling-agents-consumer-hardware-lfm2-24b-a2b)
- [ — JUN 17, 2025Press — G42 and Liquid AI Partner to Deliver Private, Local and Efficient AI Solutions to Enterprises at Scale](https://www.liquid.ai/press/g42-and-liquid-ai-partner-to-deliver-private-local-and-efficient-ai-solutions-to-enterprises-at-scale)

## Get started with Liquid AI.

Talk with our team about your use case, your data, and where it needs to run.

[Contact us](https://www.liquid.ai/connect?topic=enterprise)
