# Liquid AI ships a 350M encoder built for CPUs

> The compact LFM2.5 encoder targets fast, long-context text embeddings without a GPU.

Published by The Open Weights on Jul 27, 2026. Canonical: https://theopenweights.com/news/lfm2-5-encoder-350m-t23r

## Key facts

- Company: LiquidAI
- Model: LFM2.5-Encoder-350M
- Version: 2.5
- Category: Embeddings
- Modalities: Embeddings
- License: Apache 2.0 (Open weights, commercial use allowed)
- Parameters: 350M
- Architecture: Lfm2BidirectionalForMaskedLM
- Precision: FP32
- Weights size: 1.4 GB
- Significance: notable
- Published: 2026-07-27
- Last verified: 2026-08-04
- Hugging Face: https://huggingface.co/LiquidAI/LFM2.5-Encoder-350M
- Canonical URL: https://theopenweights.com/news/lfm2-5-encoder-350m-t23r

Liquid AI has released [LFM2.5-Encoder-350M](https://huggingface.co/LiquidAI/LFM2.5-Encoder-350M), a compact bidirectional text encoder aimed at embedding workloads that need to run quickly and cheaply. At roughly 350 million parameters, it slots into the sub-1B tier where the priority is throughput and deployability rather than raw scale.

The model is a classic encoder — bidirectional attention that reads an entire sequence at once — rather than a generative decoder. That design makes it well suited to retrieval, classification, and semantic search, where the goal is to turn text into vectors rather than produce new tokens. Liquid AI frames the release around fast long-context inference on CPU, a nod to teams that want to avoid the cost and complexity of GPU serving.

## Why it matters

Much of the open-weights conversation is dominated by ever-larger chat models, but embeddings quietly power a huge share of production AI: search, recommendations, deduplication, and retrieval-augmented generation. A small encoder that performs well on commodity hardware lowers the barrier for those pipelines.

- Compact footprint at about 350M parameters
- Bidirectional encoder architecture for embedding tasks
- Tuned for long-context inference on CPU
- Released under the permissive Apache 2.0 license

The Apache 2.0 license is notable in its own right, giving companies room to deploy and modify the model commercially. Full details and usage guidance are in Liquid AI's accompanying [write-up](https://huggingface.co/blog/LiquidAI/lfm2-5-encoders).

## Get the model

- [Hugging Face](https://huggingface.co/LiquidAI/LFM2.5-Encoder-350M)

## Sources

- [LiquidAI/LFM2.5-Encoder-350M](https://huggingface.co/LiquidAI/LFM2.5-Encoder-350M) — Hugging Face, Jul 27, 2026

---
Source: The Open Weights (https://theopenweights.com/). Aggregated and written by Claude, curated by humans. Cite as: "Liquid AI ships a 350M encoder built for CPUs", The Open Weights, Jul 27, 2026, https://theopenweights.com/news/lfm2-5-encoder-350m-t23r