PerplexitySpecialist modelOpen weights

pplx-embed-v2-context-0.6b

Summary

Perplexity announced pplx-embed-v2-context with open 0.6B and 9B weights on November 3, 2026. The multilingual vision-language model supports retrieval across text, images, and visual documents, including PDF pages.

Contextual mode embeds text chunks with the rest of the document in view. Text also has a non-contextual mode, and image pages can be split into horizontal chunks. The 0.6B checkpoint produces int8 vectors with 1,024 dimensions and shares this embedding space with the 9B model.

The model card recommends encode_queries for searches and encode for documents because they use separate trained prefixes. The weights are public, but the card states no license for this checkpoint. The reviewed sources do not establish a hosted API or price.

Similar models

More models of the same kind as pplx-embed-v2-context-0.6b, chosen by model line, developer, size, price and release date.

All specifications

Architecture

Model class: Multimodal contextual embedding model for search and retrieval.; Source: Model class
Checkpoint parameters: 595,370,176 according to Safetensors metadata.; Source: Checkpoint parameters

Input and output

Modalities: Text, images, and visual documents. Supports contextual and non-contextual text inputs as well as contextual image chunks.; Source: Modalities

Watermarking and provenance

AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.

AI watermarking: Unknown; no source
Watermarking technique: Not known; no source

Licence and access

License: No license is stated for this checkpoint in its model card.; Source: License
Shutdown: None announced; no source

Further details

Embedding output: One 1,024-dimensional int8 vector per document or document chunk. Use encode_queries for queries and encode for documents.; Source: Embedding output
Shared embedding space: Shares its 1,024-dimensional embedding space with the 9B variant.; Source: Shared embedding space

Sources and data date

Type	Evidence and data date
Additional source	Perplexity pplx-embed-v2-context-0.6b model card (retrieved November 7, 2026) · Perplexity pplx-embed-v2-context-0.6b model card · Editorial description reviewed November 7, 2026
Additional source	Perplexity pplx-embed-v2-context-0.6b weight metadata (retrieved November 7, 2026)
Additional source	Perplexity pplx-embed-v2-context-9b model card (retrieved November 7, 2026) · Perplexity pplx-embed-v2-context-9b model card · Editorial description reviewed November 7, 2026
Additional source	Perplexity v2 context announcement (retrieved November 7, 2026) · Perplexity v2 context announcement · Editorial description reviewed November 7, 2026
Additional source	pplx-embed-v2-context technical report (retrieved November 7, 2026) · pplx-embed-v2-context technical report · Editorial description reviewed November 7, 2026