PulseAugur
EN
LIVE 08:54:57

Perplexity releases contextual embedding model for RAG pipelines

By PulseAugur Editorial · [1 sources] ·

Perplexity Research and turbopuffer have released pplx-embed-v2-context-9b-preview, a new contextual embedding model designed for retrieval-augmented generation (RAG) pipelines. This model differs from traditional approaches by training to retrieve not just the answer, but also the supporting evidence needed for verification, moving away from the single 'gold passage' method. The model is available as a self-hosted preview with weights on Hugging Face under the MIT license, and it utilizes a novel training process involving a query-aware context compression model as a teacher. AI

IMPACT This new model could improve the accuracy and explainability of RAG systems by providing supporting evidence alongside answers.

RANK_REASON The release of a new embedding model with a novel training approach for RAG pipelines. [lever_c_demoted from research: ic=1 ai=1.0]

Read on MarkTechPost →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Perplexity releases contextual embedding model for RAG pipelines

How we ranked this

Signal score
11 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The release of a new embedding model with a novel training approach for RAG pipelines. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    Perplexity Releases pplx-embed-v2-context-9b-preview: A Contextual Embedding Model That Retrieves Answers and Their Supporting Evidence

    <p>Perplexity Research and turbopuffer have released pplx-embed-v2-context-9b-preview, a contextual embedding model for RAG pipelines. Each chunk is embedded with the full document in view. The real change is the training signal. The model learns to retrieve the answer along with…

RELATED ENTITIES

RELATED TOPICS