EximiusLabs has published the Fusion Embedding 1.2B preview on Hugging Face and presented it through Show HN. The submission claims that the model surpasses Gemini Embedding 2 while training only 16 million parameters, and describes the weights as open. However, the supplied evidence does not include benchmark tables, evaluation protocols, datasets, or reproducibility details. Hacker News shows a score of 7 and zero comments at publication, so the claim has received little public scrutiny so far.
No heat snapshots are available in the last 24 hours.