F2LLM-v2-14B

F2LLM-v2-14B is the largest model in CodeFuse's F2LLM-v2 family of multilingual text embedding models, described in arXiv 2603.19223. The family spans 8 sizes from 80M to 14B, trained on 60 million curated samples across more than 200 languages with emphasis on underserved mid- and low-resource languages, plus 40+ programming languages.

The project is notable for full transparency: weights, training data, training code, and intermediate checkpoints are all released under Apache 2.0. The family reports state-of-the-art MTEB results across Code, European, Scandinavian, Japanese, Vietnamese, Thai, Indic, and Persian tracks. The 14B instruct variant is the top performer in the family.