Perplexity Releases Dual-Size Embedding Models
Perplexity open-sources pplx-embed-v2-late, enabling small models to query large indexes for OCR-free visual document retrieval.
ImportanceMaterialEvidenceE2 unreplicatedWrite-upQuick
Perplexity released the pplx-embed-v2-late series of multimodal embedding models on October 7, featuring 0.6B and 9B versions under the MIT license.
Built on the Qwen3.5 architecture, these models share a unified embedding space after distillation from an 18B teacher. This allows developers to build high-quality document indexes in the cloud using the 9B model while querying them locally or on edge devices with the lightweight 0.6B model. The system directly searches rendered PDF pages without requiring OCR.
Self-reported benchmarks show the 9B model achieving 92.4% accuracy on MADQA and 64.7% nDCG@10 on ViDoRe v3 Markdown. All performance metrics are vendor-provided and have not yet been independently reproduced.