Pinecone diagnosis: recall worse than expected, isolate model vs index

TL;DR: Symptom: queries miss results you expect. Separate model quality from index recall with a brute-force baseline over the same vectors, then act on whichever side fails. Confirm it Follow the docs' recall-debug procedure: 1. Build an eval set: at least 10 queries over a 100k-vector source set, with labeled right answers.

The error

queries miss results you expect. Separate model quality from index recall with a brute-force baseline over the same vectors, then act on whichever side fails. Confirm it Follow the docs' recall-debug procedure: 1. Build an eval set: at least 10 queries over a 100k-vector source set, with labeled rig

The fix

TL;DR: Symptom: queries miss results you expect. Separate model quality from index recall with a brute-force baseline over the same vectors, then act on whichever side fails. Confirm it Follow the docs' recall-debug procedure: 1. Build an eval set: at least 10 queries over a 100k-vector source set, with labeled right answers.

The fix

1. Build an eval set: at least 10 queries over a 100k-vector source set, with labeled right answers.
2. Embed queries and corpus with your model; run brute-force search (FAISS or numpy) and score against the labels. If brute force scores badly, it is a model problem.
3. Upsert the same vectors to Pinecone, run the same queries, and compare Pinecone's returned ids to the brute-force ids. Pinecone recall near 0.99 on s1/p1 means the index is fine and the model is the issue.
4. Model problem: change the embedding model (and rebuild the index at the new dimension).
5. Index problem: check top_k (too small clips recall), filters (too strict), and pod size; escalate to support with the reproducible dataset if recall stays low.

When to use this

• This covers exactly what the title says: Pinecone diagnosis.
• You are setting this up for the first time, or auditing an existing setup.
• You want the key gotchas in one place before you start.

When not to use this

• You are doing a different workflow with Pinecone; these steps are specific to the title above.
• You need the full reference docs; this is the short path, not the manual.

Compatibility

• Mentions 0.99 in the original; behavior may differ on much older or newer releases.

Pinecone's debug-model-vs-pinecone-recall troubleshooting page prescribes this exact six-step procedure: eval framework, brute-force baseline, Pinecone comparison, ~0.99 expected recall on s1/p1, escalate with reproducible data. Verified against live docs 2026-09-26.