Weaviate
Open-source vector database, managed in Weaviate Cloud.
Overall score
55.0/100
weaviate-cloud-rest-v1
Measured Oct 5, 2026
Status
Ranked
Complete benchmark. `overall` is a weighted average of the applicable criteria, comparable to other surfaces in the same category and the same rubric version.
Coverage
86% of the rubric
Surfaces
evaluated.
We score a surface, not a product. One product often exposes several, and they are not worth the same.
weaviate-cloud-rest-v1restprimary
86% coverage
55.0/100
Criterion
by criterion.
Measured on weaviate-cloud-rest-v1, rubric v0.2.0. A criterion with no score is not counted as zero: it is excluded and the weights are renormalised.
Offload Valuejudged · weight 15 · Oct 5, 2026
78/100
The provided text is largely a navigation stub with no endpoint-level detail, so the evidence of offload is thin. What is visible — a vector database with agentic search over collections and persistent memory for LLM agents — does imply genuine offload: embedding storage and approximate nearest-neighbor retrieval over large corpora is computation an agent cannot replicate in context, and retrieval of proprietary stored vectors saves substantial reasoning tokens. However, much of the surface is still storage/CRUD over objects and schemas, and the documentation shown demonstrates no irreversible side effect (no payments, signing, sending). Substantial but not ruinous-to-replicate.
- https://docs.weaviate.io/weaviate/api/rest
Query Agent Run agentic search over your Weaviate Cloud collections
- https://docs.weaviate.io/weaviate/api/rest
Query Agent Run agentic search over your Weaviate Cloud collections Weaviate Cloud Manage and scale Weaviate in the cloud Engram Persistent memory for LLM agents and applications
- https://docs.weaviate.io/weaviate/api/rest
Query Agent Run agentic search over your Weaviate Cloud collections Weaviate Cloud Manage and scale Weaviate in the cloud Engram Persistent memory for LLM agents and applications
Interaction Costmeasured · weight 15 · Oct 5, 2026
17/100
Median over 2 successful run(s): 52349.5 tokens, 8 calls. Ratio to the best in "vector-database" (3 surfaces measured, v0.2 scale). Cold, without documentation: 1/1 succeeded.
Error Recoverabilityprobed · weight 15 · Oct 5, 2026
40/100
unknown_endpoint → HTTP 404, clarity 25/100 (field named: no, allowed values: no, RFC 9457: no) · wrong_type → HTTP 429, clarity 30/100 (field named: no, allowed values: no, RFC 9457: no) · hallucinated_enum → HTTP 422, clarity 75/100 (field named: yes, allowed values: yes, RFC 9457: no) · unknown_field → HTTP 429, clarity 30/100 (field named: no, allowed values: no, RFC 9457: no)
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema
HTTP 401: {"code":401,"message": "anonymous access not enabled. Please authenticate through one of the available methods: [API-keys, OIDC]" }
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema/nonexistent_resource_xyz
HTTP 404:
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema
HTTP 429: {"errorCode":"USAGE_LIMIT_EXCEEDED","limit":"collections","message":"collections count limit of 1 reached for this instance.","value":1}
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema
HTTP 422: {"allowed":null,"error":[{"message":"class.VectorIndexConfig can not parse: parse vector index config: invalid hfresh config: unsupported distance type 'euclidean', HFresh only supports 'cosine' or 'l
Doc Legibilitymeasured · weight 12 · Oct 5, 2026
51/100
text/markup ratio 5.9% (weight 30) · rendering without JavaScript undetermined (excluded from the calculation) (weight 25) · machine-readable spec: https://docs.weaviate.io/openapi.json (weight 25) · llms.txt: https://weaviate.io/llms.txt (weight 10) · 0 runnable example(s) (weight 10). Weighted average over 75 points of available sub-criteria.
Auth Frictionmeasured · weight 10 · Oct 5, 2026
95/100
Authorization: Bearer <key>, verified by a successful call. One header, no ceremony.
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema
HTTP 401: {"code":401,"message": "anonymous access not enabled. Please authenticate through one of the available methods: [API-keys, OIDC]" }
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema/nonexistent_resource_xyz
HTTP 404:
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema
HTTP 429: {"errorCode":"USAGE_LIMIT_EXCEEDED","limit":"collections","message":"collections count limit of 1 reached for this instance.","value":1}
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema
HTTP 422: {"allowed":null,"error":[{"message":"class.VectorIndexConfig can not parse: parse vector index config: invalid hfresh config: unsupported distance type 'euclidean', HFresh only supports 'cosine' or 'l
Safety & Reversibilityprobed · weight 10
—/100
If I retry after a timeout, do I charge the customer twice?
Payload Efficiencyprobed · weight 8 · Oct 5, 2026
52/100
Field selection via ?fields=: 822 → 15 bytes. (weight 35) · No cursor pagination marker in the response. (weight 20) · Nested envelope or non-list shape. (weight 10) · No ETag. (Compression is not measurable: the runtime decompresses it and strips the header.) (weight 10). Weighted average over 75 points of measured sub-criteria.
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema
HTTP 401: {"code":401,"message": "anonymous access not enabled. Please authenticate through one of the available methods: [API-keys, OIDC]" }
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema/nonexistent_resource_xyz
HTTP 404:
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema
HTTP 429: {"errorCode":"USAGE_LIMIT_EXCEEDED","limit":"collections","message":"collections count limit of 1 reached for this instance.","value":1}
- https://gmpqkvkaqf2kulvxfiel3a.c0.eu-central-1.aws.weaviate.cloud/v1/schema
HTTP 422: {"allowed":null,"error":[{"message":"class.VectorIndexConfig can not parse: parse vector index config: invalid hfresh config: unsupported distance type 'euclidean', HFresh only supports 'cosine' or 'l
Task Atomicitymeasured · weight 6 · Oct 5, 2026
38/100
Median over 2 successful run(s): 52349.5 tokens, 8 calls. Ratio to the best in "vector-database" (3 surfaces measured, v0.2 scale). Cold, without documentation: 1/1 succeeded.
Prior Knowledge Coveragemeasured · weight 5 · Oct 5, 2026
100/100
Median over 2 successful run(s): 52349.5 tokens, 8 calls. Ratio to the best in "vector-database" (3 surfaces measured, v0.2 scale). Cold, without documentation: 1/1 succeeded.
Async Compatibilityprobed · weight 2
—/100
If the operation is asynchronous, can I get the result without hosting a web server?
Tool-Native (MCP)binary · weight 2
—/100
Does the vendor publish an official, maintained MCP server?
Runs, verified
never on the agent’s word.
Every row is a real attempt at the category’s canonical task. Success is verified by reading the created resource back.
| Task | Tokens | Calls | Errors | Docs | Result |
|---|---|---|---|---|---|
| index-and-query-vectors | 61,105 | 8 | 3 | without | verified |
| index-and-query-vectors | 55,589 | 9 | 3 | with | verified |
| index-and-query-vectors | 49,110 | 7 | 2 | with | verified |
The same entry,
without the markup.
You publish this API and dispute a score? The methodology is public and every score above points to the evidence that produced it.