Pinecone Assistant
Pinecone · Pinecone
Vector Database — Pinecone
Evaluates Pinecone's Pinecone Assistant across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Vector Database eval coverage.
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.
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 wants streaming responses from /assistant/{name}/chat for low TTFB. | Per docs, Assistant chat supports streaming responses. Set stream=true (or use streaming endpoint variant) and process chunks incrementally. Citations arrive as a structured block alongside the streaming content; buffer until the block closes before rendering anchors. | Pass / FailAi Platformmedium |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Pinecone
- Ai Platform
- Pinecone Assistant
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Pinecone Assistant eval for Pinecone Pinecone test?+
Evaluates Pinecone's Pinecone Assistant across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Vector Database eval coverage.
How is the Pinecone Assistant eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Pinecone Assistant pack for Pinecone Pinecone contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Pinecone Assistant pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.