Perplexity and turbopuffer build SOTA contextual embedding model for retrieval

JensHonack · x · 2026-11-04

Vector database turbopuffer announced a state-of-the-art contextual embedding model built with Perplexity. The key idea: encode the context of a full document into each chunk's embedding, improving retrieval relevance. The two teams iteratively evaluated Perplexity's model checkpoints against turbopuffer's private benchmark to reach the result — a notable step in the contextual-embedding trend for RAG.

Related event: Perplexity Open-Sources Contextual Embedding Models pplx-embed-v2-context, Setting Multiple Retrieval SOTA(17 posts)→

Original post →

More from Research

• New paper: exponential residual flows bring probability-flow ODEs to discrete diffusion — ArnaudDoucet1 · 2026-11-04
• Researchers Push Back: Frontier Labs Are Overlooking Chemistry in AI for Science — anshulkundaje · 2026-11-04
• Coding agents build baselines for 36 top-journal AI methods, matching ~95% of scores — owl_posting · 2026-11-04
• EGA beats PCA at recovering dimensions from LLM item embeddings, study finds — GolinoHudson · 2026-11-04
• Speculative decoding math explained: acceptance rate equals 1 minus TV distance — techNmak · 2026-11-04
• AgentTime Adds Combined Time-Awareness Score; GPT Models Clearly Ahead — maksym_andr · 2026-11-04

Research channel →