MODELS
by Liquid AI
Liquid AI’s LFM2-24B-A2B is a 24B-parameter Mixture-of-Experts instruct text model in the LFM2 family, with about 2B active parameters, 32K context, and tool-calling support.
LFM2-24B-A2B is Liquid AI’s largest LFM2 Mixture-of-Experts text model. Liquid AI describes it as a hybrid-architecture sparse MoE checkpoint with 24B total parameters and about 2B active parameters per token, designed for efficient on-device deployment including laptops or a single GPU. Liquid AI documentation lists a 32K context window and tool-calling support. The official Hugging Face model card provides weights and examples for Transformers, vLLM, SGLang, and Docker. Together AI also provides an API endpoint for the model.
Use when you need a Liquid AI LFM2-family MoE instruct text model with 32K context and tool-calling support. Evaluate for laptop, AI PC, local, or single-GPU inference scenarios where the roughly 2B active-parameter MoE design is relevant. Use when you want official Hugging Face weights with examples for Transformers, vLLM, SGLang, or Docker. Use through Together AI when you want a hosted API endpoint for LiquidAI/LFM2-24B-A2B.
Last updated Jun 2, 2026
