Extraction
Structured data extraction from unstructured documents into JSON.
Summarization
Meeting transcripts, document summaries, and content distillation.
RAG & Retrieval
Context-grounded Q&A and fast multi-language document retrieval.
Translation
Low-latency bidirectional translation for short-to-medium text.
Many Nanos require specific prompting formats to work correctly. See each model’s page for usage guidelines.
LFM2.5 Nanos   Latest release
LFM2.5-Embedding-350M
350M · RetrievalDense bi-encoder for the smallest, fastest multilingual vector index.
LFM2.5-ColBERT-350M
350M · RetrievalLate-interaction retriever for higher accuracy and reranking quality.
LFM2.5-VL-1.6B-Extract
1.6B · Vision ExtractionExtract user-defined fields from images into structured JSON.
LFM2.5-VL-450M-Extract
450M · Vision ExtractionCompact image-to-JSON extraction model for edge workflows.
LFM2 Nanos
LFM2-350M-PII-Extract-JP
350M · ExtractionJapanese PII detection into structured JSON.
LFM2-2.6B-Transcript
2.6B · SummarizationPrivate, on-device meeting summarization from transcripts.
LFM2-350M-ENJP-MT
350M · TranslationNear real-time bidirectional Japanese/English translation.
LFM2-350M-Math
350M · ReasoningTiny reasoning model for math problem solving.
LFM2-1.2B-Extract
1.2B · DeprecatedUse LFM2.5-1.2B-Instruct for structured extraction.
LFM2-350M-Extract
350M · DeprecatedUse LFM2.5-350M for compact extraction workflows.
LFM2-1.2B-RAG
1.2B · DeprecatedUse LFM2.5-1.2B-Instruct with LFM2.5 retrieval models.
LFM2-ColBERT-350M
350M · DeprecatedUse LFM2.5-ColBERT-350M instead.
LFM2-1.2B-Tool
1.2B · DeprecatedUse LFM2.5-1.2B-Instruct for tool calling instead.