Multi-label classification. One forward pass. Zero completion tokens.
We fine-tuned LFM2.5-Encoder-230M and LFM2.5-Encoder-350M to score every label at once: no decoding loop, nothing to parse or repair. First in a new encoder demo series, watch it in action.
Check out the cookbook:
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
Why it matters
Classification through a decoder means parsing and repairing output. An encoder that scores every label in one pass removes that failure mode along with the completion-tokenThe chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.Full definition → cost.
