Embeddings 1400x Cheaper Than LLMs, Hybrid Search Remains Top Choice
jeremyphoward · x · 2026-08-22
Addressing whether LLMs can replace embedding models, the new paper "Embedder's Dilemma" finds that while LLMs now outperform specialized embedding models, they cost 1400x more. The author advocates sticking with embeddings (dense, sparse, multi-vector) combined with BM25 and rerankers. Listwise cross-encoders are also noted as an interesting option.
Related event: Harvard-Stanford Paper: LLMs Match Embedding Models but Cost ~1431x More(6 posts)→
More from Research
- Interpretability paper incoming: in-depth look at testing explanations on modern LLMs — burny_tech · 2026-10-16
- Better-learning models tend to be more interpretable, argues Nick Cammarata — burny_tech · 2026-10-16
- Cornell and Amazon revisit the 16x pretraining compute gap: diffusion LMs need only 3-8x — LucaAmb · 2026-10-16
- AI trained on the fraction of human intelligence that fit the page — r0ck3t23 · 2026-10-16
- Quotient Recurrent Reasoners hit 97.9% on Sudoku-Extreme with 78% fewer loops — burny_tech · 2026-10-16
- Mathematicians publish explainer for the Egyptian fraction proof in OpenAI's proofs collection — burny_tech · 2026-10-16