LFM2.5-Encoders for Fast Long-Context Inference on CPU

Models Datasets Spaces Buckets new Docs Enterprise Pricing Website Tasks HuggingChat Collections Languages Organizations Community Blog Posts Daily Papers Hardware Learn Discord Forum GitHub Solutions Team & Enterprise Hugging Face PRO Enterprise Support Inference Providers Inference Endpoints Storage Buckets Log In Sign Up Back to Articles a]:hidden"> LFM2.5-Encoders for Fast Long-Context Inference on CPU Team Article Published July 28, 2026 Upvote 23 +17 Fernando Fernandes Neto fernandofernandes Follow LiquidAI Edoardo Mosca EdoardoMosca Follow LiquidAI Maxime Labonne mlabonne Follow LiquidAI Leonie Monigatti iamleonie Follow LiquidAI Why we built a general-purpose encoder How the encoders are built Benchmark Results Inference speed on CPU and GPU LFM2.5-Encoder demos How to use and fine-tune LFM2.5-Encoders Load and run the model Fine-tuning for your task Get started with LFM2.5-Encoders Citation Today, we release two new encoder models on Hugging Face: LFM2.5-Encoder-230M and LFM2.5-Encoder-350M. They match the quality of larger models but stay fast as inputs get longer. This means you can run document-scale jobs on the hardware you already have, even on CPU.

With these, you can build intent routers, policy linters, PII detectors, and text classifiers that run cheaply, all day. See the live demos below.

Why we built a general-purpose encoder Last month we released LFM2.5-Retrievers, built for multilingual search. LFM2.5-Encoders come from the same family but serve a broader purpose. They're pre-trained with a masked-language objective, so you can fine-tune them for classification, token-level tasks, and search alike. Search is just one thing an encoder enables. That's why we built a general-purpose model instead of reusing the retrievers.

Encoders power many modern production NLP applications: classifiers, intent routers, safety filters. These jobs run all day, usually on CPU, on ever-longer inputs. BERT established this class of model, and recently ModernBERT pushed its accuracy, speed, and context further. LFM2.5-Encoders take the next step on the LFM2 architecture, where cost grows slowly as inputs grow.

How the encoders are built We initialize the encoders from their respective LFM2 decoder backbones: LFM2.5-230M and LFM2.5-350M. Then we turn each causal decoder into a bidirectional encoder with a few changes:

Benchmark Results We fine-tune each model fully on every task and report the resulting score. Across the table, that's 14 models on 17 tasks pulled from GLUE, SuperGLUE, and multilingual classification.

We report the mean across five held-out seeds, so the numbers are stable run to run. The full framework and raw results are open-sourced.

LFM2.5-Encoder-350M ranks fourth of the 14 models. The three ahead of it are all larger, including a 3.5B model nearly 10 times its size. LFM2.5-Encoder-230M beats ModernBERT-base and every EuroBERT model, while being smaller than most of them. Both also score well above our own LFM2.5-Retrievers here.

Official Announcement

Read the full update directly from the official source at Hugging Face News.

Stay tuned to Modelverse for real-time model analysis and benchmark coverage.