LFM2.5-Encoder-350M

LFM2.5-Encoder-350M is a multilingual bidirectional encoder model built on the LFM2 hybrid architecture. It is designed for high-performance, on-device tasks such as text classification, retrieval, and semantic similarity across 15 languages.

LFM2.5-Encoder-350M is a multilingual bidirectional encoder model built on the LFM2 hybrid architecture. It is designed for high-performance, on-device tasks such as text classification, retrieval, and semantic similarity across 15 languages.

• 018,192-token context window
• 02Multilingual support across 15 languages
• 03Efficient LFM2 hybrid backbone with gated convolutions
• 04Optimized for on-device CPU inference
• 05Full bidirectional attention for masked language modeling