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

What changed

Here's what you get: - Strong for their size: match or beat larger encoders on GLUE, SuperGLUE, and multilingual tasks. - 8,192-token context with latency that grows slowly as inputs get longer. Last month we released LFM2.5-Retrievers, built for multilingual search.

Why it matters

A concrete addition to Practical AI: Tools, Models & Frameworks: it changes what's available to builders today rather than being general commentary.

How it compares

Related prior coverage to compare against:

• Scaling-up BERT Inference on CPU (Part 1)
• Scaling up BERT-like model Inference on modern CPU - Part 2
• Introducing HELMET: Holistically Evaluating Long-context Language Models

• Scaling-up BERT Inference on CPU (Part 1) similarity 0.28
• Scaling up BERT-like model Inference on modern CPU - Part 2 similarity 0.23
• Introducing HELMET: Holistically Evaluating Long-context Language Models similarity 0.21