LFM2

LFM2

Liquid AI
Mu

Mu

Microsoft
+
+

Related Products

  • Google AI Studio
    41 Ratings
    Visit Website
  • LM-Kit.NET
    29 Ratings
    Visit Website
  • Gemini Enterprise Agent Platform
    999 Ratings
    Visit Website
  • Google Cloud Speech-to-Text
    366 Ratings
    Visit Website
  • Iru
    1,385 Ratings
    Visit Website
  • ManageEngine Endpoint Central
    3,286 Ratings
    Visit Website
  • DriveStrike
    24 Ratings
    Visit Website
  • optivalue.ai
    4 Ratings
    Visit Website
  • Kitecyber
    82 Ratings
    Visit Website
  • Bitdefender Ultimate Small Business Security
    5 Ratings
    Visit Website

About

LFM2 is a next-generation series of on-device foundation models built to deliver the fastest generative-AI experience across a wide range of endpoints. It employs a new hybrid architecture that achieves up to 2x faster decode and prefill performance than comparable models, and up to 3x improvements in training efficiency compared to the previous generation. These models strike an optimal balance of quality, latency, and memory for deployment on embedded systems, allowing real-time, on-device AI across smartphones, laptops, vehicles, wearables, and other endpoints, enabling millisecond inference, device resilience, and full data sovereignty. Available in three dense checkpoints (0.35 B, 0.7 B, and 1.2 B parameters), LFM2 demonstrates benchmark performance that outperforms similarly sized models in tasks such as knowledge recall, mathematics, multilingual instruction-following, and conversational dialogue evaluations.

About

Mu is a 330-million-parameter encoder–decoder language model designed to power the agent in Windows settings by mapping natural-language queries to Settings function calls, running fully on-device via NPUs at over 100 tokens per second while maintaining high accuracy. Drawing on Phi Silica optimizations, Mu’s encoder–decoder architecture reuses a fixed-length latent representation to cut computation and memory overhead, yielding 47 percent lower first-token latency and 4.7× higher decoding speed on Qualcomm Hexagon NPUs compared to similar decoder-only models. Hardware-aware tuning, including a 2/3–1/3 encoder–decoder parameter split, weight sharing between input and output embeddings, Dual LayerNorm, rotary positional embeddings, and grouped-query attention, enables fast inference at over 200 tokens per second on devices like Surface Laptop 7 and sub-500 ms response times for settings queries.

Platforms Supported

Windows Not Supported
Mac Not Supported
Linux Not Supported
Cloud Supported
On-Premises Not Supported
iPhone Not Supported
iPad Not Supported
Android Not Supported
Chromebook Not Supported

Platforms Supported

Windows Supported
Mac Not Supported
Linux Not Supported
Cloud Not Supported
On-Premises Not Supported
iPhone Not Supported
iPad Not Supported
Android Not Supported
Chromebook Not Supported

Audience

Developers and engineering teams needing a solution offering foundation models without reliance on cloud infrastructure

Audience

Developers seeking a solution to navigate and configure system settings through natural language

Support

Phone Support Supported
24/7 Live Support Not Supported
Online Supported

Support

Phone Support Supported
24/7 Live Support Not Supported
Online Supported

API

Offers API Not Supported

API

Offers API Not Supported

Screenshots and Videos

Screenshots and Videos

Pricing

No information available.
Free Version Not Supported
Free Trial Supported

Pricing

No information available.
Free Version Not Supported
Free Trial Not Supported

Reviews/Ratings

Overall 0.0 / 5
ease 0.0 / 5
features 0.0 / 5
design 0.0 / 5
support 0.0 / 5

This software hasn't been reviewed yet. Be the first to provide a review:

Review this Software

Reviews/Ratings

Overall 0.0 / 5
ease 0.0 / 5
features 0.0 / 5
design 0.0 / 5
support 0.0 / 5

This software hasn't been reviewed yet. Be the first to provide a review:

Review this Software

Training

Documentation Supported
Webinars Not Supported
Live Online Supported
In Person Not Supported

Training

Documentation Supported
Webinars Supported
Live Online Supported
In Person Supported

Company Information

Liquid AI
Founded: 2023
United States
www.liquid.ai/blog/liquid-foundation-models-v2-our-second-series-of-generative-ai-models

Company Information

Microsoft
Founded: 1975
United States
blogs.windows.com/windowsexperience/2025/06/23/introducing-mu-language-model-and-how-it-enabled-the-agent-in-windows-settings/

Alternatives

Ministral 8B

Ministral 8B

Mistral AI

Alternatives

CodeT5

CodeT5

Salesforce
Seed2.0 Lite

Seed2.0 Lite

ByteDance
GLM-OCR

GLM-OCR

Z.ai
Ministral 3B

Ministral 3B

Mistral AI
Gemma 2

Gemma 2

Google
Falcon-7B

Falcon-7B

Technology Innovation Institute (TII)

Categories

AI Models Supported

Categories

AI Models Supported

Integrations

Hugging Face Supported
OpenAI Supported
OpenRouter Supported
Qwen3 Supported
Together AI Supported

Integrations

Hugging Face Not Supported
OpenAI Not Supported
OpenRouter Not Supported
Qwen3 Not Supported
Together AI Not Supported
Claim LFM2 and update features and information
Claim LFM2 and update features and information
Claim Mu and update features and information
Claim Mu and update features and information