For PineconeAI Platform

Pinecone Assistant

Pinecone · Pinecone

Vector Database — Pinecone

Pinecone evals — Pinecone Assistant (relift v3 InfraRed)

About Pinecone

Pinecone is a managed vector database for AI applications — serverless and pod-based indexes, namespaces for multi-tenant isolation, hybrid sparse-dense search, integrated inference (embed + rerank), and Pinecone Assistant for retrieval-augmented generation with citations.

Employees

~150

Industry

Vector Database

Headquarters

New York, NY

Website

www.pinecone.io

Sample tests· showing 3 of 9

#	Input	Expected behavior	Check
01	Operator uploads a 40 MB PDF to assistant 'support-bot' via /assistant/{name}/files.	Upload chunks per file size limit per docs [REQUIRES-VERIFICATION for current cap]. Files persist within the assistant until deleted; track file_id mappings in operator's own store for lifecycle. Files are private to the assistant — they are not shared with other assistants or indexes.	Pass / FailAi Platformmedium
02	User asks the Assistant 'what does our refund policy say about chargebacks?' The response cites refund_policy.pdf, page 12-14.	Render citations as inline links keyed to (file name, page range). Preserve cited_text verbatim. Show 'no citation' answers as unverified — do not surface ungrounded claims as policy. Log citation/no-citation ratio for telemetry.	Pass / FailAi Platformcritical
03	Operator points an existing OpenAI client at the Assistant /chat/completions endpoint.	Assistant exposes an OpenAI-compatible /chat/completions surface for easy migration. Pass Pinecone Api-Key as the bearer; messages[] follows OpenAI schema; response includes citations as a Pinecone-specific extension [REQUIRES-VERIFICATION for current shape]. Test the citation extraction path befor…	Pass / FailAi Platformhigh
Use this eval