Points Upsert And Payload
Qdrant · Qdrant
Vector Database — Qdrant
Evaluates Qdrant's Points: Upsert & Payload 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 Qdrant
Qdrant is an open-source vector database and similarity-search engine — collections with configurable vector size/distance, payload filtering (must/should/must_not), named and sparse vectors, hybrid search with prefetch and RRF/DBSF fusion, scalar/product/binary quantization, and the managed Qdrant Cloud with API-key/JWT auth and payload-based multitenancy.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Agent upserts a point via PUT /collections/{name}/points with {id:42, vector:[...], payload:{title:'Widget', price:9.99}}. | Each point carries id (uint64 or UUID), vector(s) matching the collection's vector config, and an optional JSON payload. Upsert replaces the full point at that id (vector + payload) — to change only payload, use the payload endpoints instead so the vector is not required. | Pass / FailAi Platformhigh |
| 02 | Agent upserts point id=7 (integer) and later upserts id='7' (string). It expects both to address the same point. | Qdrant point ids are either unsigned integers or UUID strings — these are distinct id spaces. Integer 7 and the UUID/string '7' are NOT the same point. Pick one id scheme per collection and use it consistently; do not rely on string/int coercion. | Pass / FailAi Platformhigh |
| 03 | Agent bulk-loads 100k points using the batch form {ids:[...], vectors:[...], payloads:[...]} in chunked requests. | Use the batch upsert form with parallel arrays (ids[], vectors[], payloads[]) of equal length; chunk into reasonable request sizes and set wait=false for throughput, polling collection status for indexing completion. Verify the three arrays stay index-aligned so payloads attach to the right vectors. | Pass / FailAi Platformhigh |
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
- Qdrant
- Ai Platform
- Points Upsert And Payload
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 Points Upsert And Payload eval for Qdrant Qdrant test?+
Evaluates Qdrant's Points: Upsert & Payload 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 Points Upsert And Payload 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 Points Upsert And Payload pack for Qdrant Qdrant 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 Points Upsert And Payload 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.