Lattice Released: An 8MB Static Embedding Model That Embeds Wikipedia in Minutes
IgorCarron · x · 2026-08-08
Developer Erik Kaum released Lattice, his summer project featuring an ultra-lightweight static embedding model.
Key Highlights:
- Tiny Footprint: Only 7.94 MB on disk.
- High Speed: Capable of embedding the entire English Wikipedia in just 7 minutes on an Apple M2 MacBook Air.
- Decent Performance: Achieves a 0.530 NDCG@10 score on the NanoBEIR benchmark.
Additionally, a related team released multilingual dataset versions (mLateOn and mDenseOn) based on this work, aiming to advance tokenization and sparse model research in Information Retrieval (IR).
Related event: 8MB Static Embedding Model Lattice Processes English Wikipedia in 7 Minutes(2 posts)→
More from Research
- FBI method trains for Pass@K with one sample per update at 12.3% of the cost — EmtiyazKhan · 2026-10-16
- Berkeley researchers sample humanoid robot policies directly from weight space with diffusion — berkeley_ai · 2026-10-16
- OneWM: unified causal training shows world and action modeling improve each other — IAAI233 · 2026-10-16
- Baidu's Virtual Head Attention tames the KV cache, FLOPs, quality triangle in long-context LLMs — baidu · 2026-10-16
- 5.8M-param Speech Event Detector adds laughs, breaths and背景 sounds to Whisper transcripts — Ayush Kumar Bar · 2026-10-16
- Huawei's SUPRA-Tex turns one image into relightable 3D materials with WYSIWYG fidelity — Chen Yang · 2026-10-16