# OpenSearch vs Vespa

> Vespa scores 70 (BB) to OpenSearch's 68 (B) for vector search. Prices, MCP, x402, uptime and agent notes side by side.

- Canonical: https://www.anchorterminal.com/compare/opensearch-vs-vespa
- Markdown: https://www.anchorterminal.com/compare/opensearch-vs-vespa.md (~2,500 tokens)
- Slim: https://www.anchorterminal.com/compare/opensearch-vs-vespa.min.md (~630 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/compare/opensearch-vs-vespa.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (full text: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- Updated: 2026-10-09

Vespa scores 70 (BB) on agent readiness against OpenSearch's 68 (B), and leads in 4 of 7 scored categories. OpenSearch leads on schema & documentation and payments & pricing. Both do vector search.

- OpenSearch: grade B, 68/100, rank #233 of 950. Markdown https://www.anchorterminal.com/tools/opensearch.md · JSON https://www.anchorterminal.com/api/v1/tools/opensearch.json
- Vespa: grade BB, 70/100, rank #165 of 950. Markdown https://www.anchorterminal.com/tools/vespa.md · JSON https://www.anchorterminal.com/api/v1/tools/vespa.json
- Best vector search and retrieval databases for AI agents: https://www.anchorterminal.com/best/vector-search/index.md
- All 85 retrieval comparisons: https://www.anchorterminal.com/compare/vector-search/index.md

## Which one, for what

### OpenSearch (B)

Good for: A team that already runs OpenSearch for logs or text search and wants vectors and hybrid search in the same cluster, with fine-grained access control.

Ahead on:
- Schema & documentation, 77 against 58
- Payments & pricing, 60 against 40

Also in its favour:
- Runs on your own machine

Watch for: No hosted service from the foundation. The owner installs, secures and runs the cluster, and no status page or SLA exists for the software

### Vespa (BB)

Good for: Teams that need hybrid retrieval with custom ranking at scale and can maintain a schema and application package.

Ahead on:
- Security & auth, 75 against 70
- Maintenance & community, 91 against 84
- Transparency & trust, 85 against 57

Also in its favour:
- Agent-ready, a grade of BB or better
- Free to start without a card

Watch for: No OpenAPI file was found in the docs index. The APIs are described in reference pages only


## Score by category

| Category | Weight | OpenSearch | Vespa | Edge |
| --- | --- | --- | --- | --- |
| Reliability | 16% (20 this run) | 77 | 80 | Vespa +3 |
| Performance | 10%, pending | pending | pending | not scored in this run |
| Schema & documentation | 13% (16.2 this run) | 77 | 58 | OpenSearch +19 |
| Agent ergonomics | 13% (16.2 this run) | 80 | 80 | even |
| Security & auth | 14% (17.5 this run) | 70 | 75 | Vespa +5 |
| Payments & pricing | 10% (12.5 this run) | 60 | 40 | OpenSearch +20 |
| Task success | 10%, pending | pending | pending | not scored in this run |
| Maintenance & community | 7% (8.8 this run) | 84 | 91 | Vespa +7 |
| Transparency & trust | 7% (8.8 this run) | 57 | 85 | Vespa +28 |
| Negative events | ≤15 | -5 | -2 | |
| **Total** | | **68 · B** | **70 · BB** | |

## Facts side by side

| Fact | OpenSearch | Vespa |
| --- | --- | --- |
| Kind | HTTP API | HTTP API |
| Vendor | OpenSearch Software Foundation | Vespa.ai AS |
| Hosted endpoint | no (local only) | no (local only) |
| Transports | HTTP, stdio | HTTP |
| Auth | OAuth or key | OAuth or key |
| Pricing | Free | Pay per use |
| x402 | no | no |
| Licence | Apache-2.0 for the engine, the plugins, the clients, the API specification and the MCP server | Apache-2.0 for the engine, the Vespa CLI and pyvespa. Vespa Cloud is a paid hosted service |
| Tools exposed | 9 | none |
| Read-only variant documented | no | no |
| llms.txt | no | yes |
| Last release | 2026-09-22 | 2026-10-01 |
| Terms last updated | no document linked | no document linked |
| Privacy policy last updated | no document linked | no date given |
| Customer content may train models |  |  |
| Terms restrict automated access |  |  |
| Terms restrict benchmarking |  |  |
| Terms or service can change without notice |  |  |
| Arbitration or class-action waiver |  |  |
| Popularity | 14k stars | 7.1k stars, 180k PyPI/wk |

## Verdicts

**OpenSearch.** OpenSearch 3.9.0 runs vector, text and hybrid search under Apache-2.0, with an OpenAPI 3.1 description of its REST API and API keys limited by permission and index pattern. The owner runs the cluster, no status page or SLA comes with the software, and no tool in the MCP server sets a read-only or destructive annotation.

**Vespa.** Vespa Cloud publishes hourly unit prices, a 99.5% uptime agreement for paying customers and a trial with $300 of credits and no card. Every call needs an application package deployed first, no OpenAPI file was found in the reviewed documentation, and Vespa.ai says it has no official public MCP server.

## Before you call either

### OpenSearch

1. Create the index with `index.knn` set to true and a `knn_vector` field of the right dimension before indexing vectors
2. Put the `filter` clause inside the `knn` query for filtering during the search. A filter outside it runs afterwards and can return fewer than `k` hits
3. Ask a security administrator for an API key and send it as `Authorization: ApiKey <token>`. The token is shown once and lasts 90 days at most
4. Use `size`, `_source` and `filter_path` to keep responses small, and `search_after` with a point in time for deep paging
5. Set `OPENSEARCH_SETTINGS_ALLOW_WRITE=false` on the MCP server unless writes are needed. `GenericOpenSearchApiTool` can call any endpoint

### Vespa

1. Deploy an application package with a schema before feeding or querying. Use `vespa deploy` and wait for the endpoint to report ready
2. Send a token as `Authorization: Bearer` only to the endpoint marked Token. The mTLS endpoint needs the client certificate and key
3. Set `hits`, `offset` and `presentation.summary` on queries. `hits` is capped at 400 by default and the query timeout defaults to 0.5s
4. On 429 from `/document/v1` reduce the feed rate and back off. On 507 stop feeding, because the cluster is out of memory or disk
5. Use the `condition` parameter for test-and-set writes and expect 412 when it fails or the document is missing

## Questions

### Which is better for AI agents, OpenSearch or Vespa?

Vespa scores 70 (BB) on agent readiness against OpenSearch's 68 (B), and leads in 4 of 7 scored categories. OpenSearch leads on schema & documentation and payments & pricing.

### Do OpenSearch and Vespa need an API key?

Both take an API key or an OAuth sign-in.

### Can an agent call OpenSearch and Vespa without installing anything?

OpenSearch runs on your own machine, with no hosted endpoint listed. No hosted endpoint is listed for Vespa.

### Are OpenSearch and Vespa open source?

Yes. OpenSearch is open source (Apache-2.0 for the engine, the plugins, the clients, the API specification and the MCP server). Vespa is open source (Apache-2.0 for the engine, the Vespa CLI and pyvespa. Vespa Cloud is a paid hosted service).


## For agents

- This comparison as JSON: https://www.anchorterminal.com/compare/opensearch-vs-vespa.json, and with the fewest tokens: https://www.anchorterminal.com/compare/opensearch-vs-vespa.min.md
- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {"a": "opensearch", "b": "vespa"}`. From a terminal: `anchor compare opensearch vespa`
- Each listing in full: https://www.anchorterminal.com/api/v1/tools/opensearch.json and https://www.anchorterminal.com/api/v1/tools/vespa.json

## Other comparisons with OpenSearch or Vespa

- [Chroma API + MCP vs OpenSearch](https://www.anchorterminal.com/compare/chroma-vs-opensearch.md)
- [Chroma API + MCP vs Vespa](https://www.anchorterminal.com/compare/chroma-vs-vespa.md)
- [Elasticsearch vs OpenSearch](https://www.anchorterminal.com/compare/elasticsearch-vs-opensearch.md)
- [Elasticsearch vs Vespa](https://www.anchorterminal.com/compare/elasticsearch-vs-vespa.md)
- [Epsilla Vector Database vs OpenSearch](https://www.anchorterminal.com/compare/epsilla-vs-opensearch.md)
- [Epsilla Vector Database vs Vespa](https://www.anchorterminal.com/compare/epsilla-vs-vespa.md)
- [LanceDB vs OpenSearch](https://www.anchorterminal.com/compare/lancedb-vs-opensearch.md)
- [LanceDB vs Vespa](https://www.anchorterminal.com/compare/lancedb-vs-vespa.md)
- [Meilisearch vs OpenSearch](https://www.anchorterminal.com/compare/meilisearch-vs-opensearch.md)
- [Meilisearch vs Vespa](https://www.anchorterminal.com/compare/meilisearch-vs-vespa.md)
- [Milvus and Zilliz Cloud API + MCP vs OpenSearch](https://www.anchorterminal.com/compare/milvus-zilliz-vs-opensearch.md)
- [Milvus and Zilliz Cloud API + MCP vs Vespa](https://www.anchorterminal.com/compare/milvus-zilliz-vs-vespa.md)
- [OpenSearch vs Pinecone API + MCP](https://www.anchorterminal.com/compare/opensearch-vs-pinecone.md)
- [OpenSearch vs Qdrant API + MCP](https://www.anchorterminal.com/compare/opensearch-vs-qdrant.md)
- [OpenSearch vs Typesense API + MCP](https://www.anchorterminal.com/compare/opensearch-vs-typesense.md)
- [OpenSearch vs Upstash Vector API + MCP](https://www.anchorterminal.com/compare/opensearch-vs-upstash-vector.md)
- [OpenSearch vs Weaviate API + MCP](https://www.anchorterminal.com/compare/opensearch-vs-weaviate.md)
- [Pinecone API + MCP vs Vespa](https://www.anchorterminal.com/compare/pinecone-vs-vespa.md)
- [Qdrant API + MCP vs Vespa](https://www.anchorterminal.com/compare/qdrant-vs-vespa.md)
- [Typesense API + MCP vs Vespa](https://www.anchorterminal.com/compare/typesense-vs-vespa.md)
- [Upstash Vector API + MCP vs Vespa](https://www.anchorterminal.com/compare/upstash-vector-vs-vespa.md)
- [Vespa vs Weaviate API + MCP](https://www.anchorterminal.com/compare/vespa-vs-weaviate.md)