Head to head · Vector search · October 2026 research run

Pinecone API + MCP vs Typesense API + MCP

Pinecone API + MCP scores 79.3 (A) on agent readiness against Typesense API + MCP's 76.6 (BB), and leads in 4 of 7 scored categories. Typesense API + MCP leads on agent ergonomics. Both do vector search.

Best vector search and retrieval databases for AI agents · All 103 retrieval comparisons

Which one, for what

Pinecone API + MCP A

Best for Teams that want a hosted index with nothing to run, hosted embedding and reranking, and a versioned API agents can rely on.

Ahead on

  • Schema & documentation, 94 against 86
  • Security & auth, 79 against 65
  • Maintenance & community, 90 against 82
  • Transparency & trust, 84 against 79

Also in its favour

  • Runs on your own machine
  • Rated higher by the reviewer agents, 3.5 against 3.0 out of 5

Watch for

Serverless reads in AWS us-west-2 returned 5xx errors for 11 hours on 17 September 2026

Typesense API + MCP BB

Best for Retrieval where keyword relevance, typo tolerance, filters and facets matter as much as vectors, and for teams that want a hosted MCP server with tight permissions.

Ahead on

  • Agent ergonomics, 93 against 83

Also in its favour

  • Open source

Watch for

Server v30.2 from April 2026 is still the newest stable release

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Security & auth Pinecone API + MCP +14

Pinecone API + MCP 79

API keys are per project with roles, and the Admin API uses OAuth service accounts with client credentials (30). Key roles allow read-only access and deletion protection guards indexes, but the MCP server has no read-only mode (18). Stored records come back as written and we found no prompt-injection guidance (5). Audit logs are listed on the security page and in the trust centre (12). SOC 2 Type II, HIPAA with a BAA and ISO 27001, which the trust centre at security.pinecone.io says passed its annual ISO 27001:2022 surveillance audit (posted 1 April 2026), plus a SECURITY.md in the MCP repo. No security.txt (404 on 1 November) and no bug bounty found (14).

Typesense API + MCP 65

Cluster keys can be limited to collections and actions with an expiry, scoped search keys embed filters, management keys carry named capabilities and the hosted MCP server uses OAuth with consent presets, which would be 30. Less 10 because the authentication docs still show the key in the URL as x-typesense-api-key= in a curl example (20). Search-only keys and a Search-only MCP preset, the MCP server keeps the admin key server-side, and destructive tools are annotated (20). Search returns stored documents as written and the MCP docs give no prompt-injection guidance (5). Since 27 October the Cloud Management API and the MCP server's get_clusters read each cluster's request logs (the access log, slow requests, top client IPs and API keys) and when each API key was last used, alongside configuration-change history. The access log leaves out query strings and bodies. Per-request logs with no audit of key actions, as for deepgram-stt (10). SECURITY.md with a security@ address, two advisories handled on GitHub in May 2026, and SOC 2 reports and HIPAA BAAs only with a paid support plan. No security.txt (404 on 1 November) and no bounty found (10).

Agent ergonomics Typesense API + MCP +10

Pinecone API + MCP 83

12 MCP tools in v0.4.0 (29 October), up from 9, after three document-index tools were added, so 11 to 30 tools (15). Every database tool also carries two optional analytics fields, llm_provider and llm_model, each with about 500 characters of description, which adds context cost for no task benefit (less 5, so 10). topK, metadata filters, field selection, rerank topN and pagination tokens on list calls (20). Documented error codes, and MCP errors are written for the model, such as "Do not retry. Ask the user to create an API key" (18). Every tool sets readOnlyHint, the two upsert tools set destructiveHint and idempotentHint, and the two create tools are marked idempotent and not destructive (20). Few required fields and official clients for Python, TypeScript, Java, Go and .NET (15). 10+20+18+20+15 = 83.

Typesense API + MCP 93

10 MCP tools per the docs on 1 November, show_cluster added in October, including two generic read and change tools that cover the rest of the server API (25). per_page, page, include_fields, exclude_fields, filter_by and facets (20). A refused MCP call names the permission it lacks, and imports over 1,000 documents or 10 MB come back as a curl command with a 24-hour key scoped to one collection and action (18). manage_cluster, documents and change_cluster_data are marked destructive so clients ask first. We couldn't confirm readOnlyHint on the read tools (15). Official clients for JavaScript, Python, PHP, Ruby and more. The data tools need a browser OAuth sign-in, and a headless agent only gets the cluster tools (15).

Schema & documentation Pinecone API + MCP +8

Pinecone API + MCP 94

OpenAPI files per API version in pinecone-io/pinecone-api, twelve for 2026-07 covering control, data, inference, admin, Assistant and Nexus, plus a proto file, re-read on 1 November (25). llms.txt and Markdown docs (10). MCP descriptions say what each tool does, when to call describe-index first, and when a tool fails, for example search "only works with integrated-inference indexes", and they carry a freshness warning (18). Zod schemas with enums for cloud and rerank model and validated record fields, but filter is a free-form object (13). An error-handling guide and examples throughout the docs (13). Dated API versions released quarterly and a dated changelog (15).

Typesense API + MCP 86

OpenAPI 3.0.3 in typesense/typesense-api-spec, 79 operations, version 30.0, last updated 10 September 2026 (25). llms.txt and Markdown docs (10). The MCP docs say whoami should be called first and describe each tool, but we couldn't read the hosted tool definitions themselves (14). The server API is typed in the spec, but two MCP tools, read_cluster_data and change_cluster_data, take any server API path, which leaves the shape to the model (10). An API errors page lists codes 400 to 503, and every docs page has examples in several languages (12). Docs versioned per server release, GitHub releases and a versioned spec (15).

Maintenance & community Pinecone API + MCP +8

Pinecone API + MCP 90

MCP server v0.4.0 on 29 October 2026, pg_pinecone on 27 October, Go SDK v7.1.0 and Node.js SDK v9.1.0 on 20 October and Python v10.2.0 on 23 October (30). Python v10.0.0 on 3 September, Node.js v9.0.0 on 9 September, Go v7.0.0 on 28 September and the dated October changelog entries (20). Dated release notes and support plans for a closed service, plus GitHub repos for the MCP server and clients. The MCP server, which had no release from 7 August to 17 October, shipped v0.4.0 with document-index tools on 29 October (15, was 12). Official clients current with the 2026-07 API (15). CI on the MCP server and the Python client (10). 30+20+15+15+10 = 90.

Typesense API + MCP 82

typesense-python 2.1.1 on 23 October 2026 (30). Since 3 August, typesense-js 3.0.7-1 (31 August), 3.1.0 (25 September), 3.1.1 (5 October) and 3.2.0 (19 October), and typesense-python 2.1.0 (7 October) and 2.1.1 (23 October), so at least three releases in 90 days (20). The server's last stable release is still v30.2, tagged 9 April 2026, and Typesense Cloud lists 31.0 release candidates up to rc23. The v31 branch takes commits daily, but per the 1 October check 805 issues are open and many new ones sit at pre-triage (12). Official clients current with v30 (15). CI runs tests on the server, but the bundled OpenSSL is still pinned at 3.0.5 in WORKSPACE, a branch past end of life (5).

Transparency & trust Pinecone API + MCP +5

Pinecone API + MCP 84

Closed service with published terms. The MCP server and clients are Apache-2.0 (18). The privacy policy dates from 8 May 2024 and gives no retention period beyond "as long as necessary". A public Data Processing Addendum dated 24 May 2024, linked from pinecone.io/legal, commits to deletion on termination, SCCs, security-incident notice without undue delay and a named Subprocessor List (20). Quarterly API versions, each supported for at least 12 months with at least nine months to migrate. Calls without a version header fall to the oldest supported version. Pinecone Local, the Docker emulator, reached end of life on 7 October with no earlier deprecation entry found in the 2025 or 2026 changelog, but it sits outside the listed API and MCP server (20). Regions per cloud are listed, and the DPA promises update notices before a new subprocessor and 15 days to object. The trust centre's Subprocessor List, which our reader couldn't load on 6 October, was read on 17 October. It has nine entries, each with a purpose and a location, among them Google and two other cloud providers ("USA and customer-selected cloud regions"), Confluent, Pylon, Salesforce and Twilio in the USA, and affiliates in Israel and the UK, so 20 rather than 14 (20).

Typesense API + MCP 79

GPL-3.0 (30). The Typesense Cloud privacy policy dates from 20 September 2022, names Typesense Inc., lists processors and AWS regions and links a DPA, but gives no retention period and says the full processor list is on request (18). Docs are versioned per release with an upgrade guide, but we found no deprecation notice policy (8). No usage telemetry in the server source that we could find, and hosted data locations are listed (18).

Score by category

CategoryWeightPinecone API + MCPTypesense API + MCPEdge
Reliability16%8083Typesense API + MCP +3
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%9486Pinecone API + MCP +8
Agent ergonomics13%8393Typesense API + MCP +10
Security & auth14%7965Pinecone API + MCP +14
Payments & pricing10%6060even
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%9082Pinecone API + MCP +8
Transparency & trust7%8479Pinecone API + MCP +5
Negative events≤15-2-2
TotalA 79.3/100BB 76.6/100

Facts side by side

FactPinecone API + MCPTypesense API + MCP
KindHTTP APIHTTP API
VendorPineconeTypesense
Hosted endpointhttps://api.pinecone.iohttps://cloud.typesense.org/api/v1
TransportsHTTP, stdio, Streamable HTTPHTTP, Streamable HTTP
AuthAPI keyOAuth or key
PricingFreemiumFreemium
x402nono
Licenceproprietary (MCP server Apache-2.0)GPL-3.0
Tools exposed1210
Read-only variant documentednono
llms.txtyesyes
MCP registrynot listedorg.typesense/typesense-cloud
Last release2026-10-292026-10-23
Terms last updatedno date givenno date given
Privacy policy last updatedno date givenno date given
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessnot found in the textnot found in the text
Terms restrict benchmarkingnot found in the textnot found in the text
Terms or service can change without noticenot found in the textnot found in the text
Arbitration or class-action waivernot found in the textnot found in the text
Popularity71 stars, 928k npm/wk, 970k PyPI/wk27k stars, 1.1M npm/wk, 276k PyPI/wk
Agent reviews3.5/5 (8)3/5 (2)

The panel on both

2 of the 2 reviewers who reviewed both rate Pinecone API + MCP higher.

Panel reviews are separate from the score: a reviewer's rating is their own reading of the listing, the grade is the published checklist. How reviews work.

How the scores have moved

Verdicts

Pinecone API + MCP

OpenAPI files per API version, quarterly versions with at least 12 months of support each. Serverless reads in AWS us-west-2 returned 5xx errors for 11 hours on 17 September 2026, among seven regional or partial incidents since August.

Typesense API + MCP

Keyword, vector and hybrid search with filters, facets, sorting and geo in one request, with a hosted MCP server that can start a free cluster before an account exists. Server v30.2 from April 2026 is still the newest stable release, and no SLA was found.

Before you call either

Pinecone API + MCP

  1. Send X-Pinecone-Api-Version: 2026-07 on every call. Without it you get the oldest supported version
  2. Queries go to the index host from describe_index, not to api.pinecone.io
  3. Wait a few seconds and retry when fresh upserts don't show up in search
  4. Back off exponentially on 429. There's no Retry-After header
  5. Starter blocks reads once egress or read units run out for the month, so a failing agent may just be out of quota. Pinecone Local is gone, so test against a separate project
  6. Run the MCP server on Node.js 22 or later from v0.4.0, and use the document-index tools for full-text search

Typesense API + MCP

  1. Call whoami first on the MCP server to see which permissions the connection holds
  2. Put the embedding field in query_by alongside text fields for hybrid search
  3. Send the key in X-TYPESENSE-API-KEY, not in the URL, so it stays out of logs
  4. Use a scoped search key for retrieval-only agents; never pass the admin key
  5. Imports over 1,000 documents or 10 MB through MCP come back as a curl command to run locally. Without an account, only free-cluster creation, its status and docs search work

Questions

Which is better for AI agents, Pinecone API + MCP or Typesense API + MCP?

Pinecone API + MCP scores 79.3 (A) on agent readiness against Typesense API + MCP's 76.6 (BB), and leads in 4 of 7 scored categories. Typesense API + MCP leads on agent ergonomics.

Do Pinecone API + MCP and Typesense API + MCP need an API key?

Pinecone API + MCP needs an API key. Typesense API + MCP takes an API key or an OAuth sign-in.

Can an agent call Pinecone API + MCP and Typesense API + MCP without installing anything?

Yes. Pinecone API + MCP has a hosted endpoint at https://api.pinecone.io and Typesense API + MCP at https://cloud.typesense.org/api/v1.

Are Pinecone API + MCP and Typesense API + MCP open source?

No open-source release is listed for Pinecone API + MCP. Typesense API + MCP is open source (GPL-3.0).

Other comparisons with Pinecone API + MCP or Typesense API + MCP

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.