Rank #2 of 7 in Vector Databases & Memory Stores

Milvus

Name: Milvus
Author: LF AI & Data Foundation / Zilliz

Open Source

LF AI & Data Foundation / Zilliz

milvus.io ↗Try it →

PA Score36/100 AGENT-READY58/100 BUILT-IN AI10/100 API10/100

MCP✓CLI—API✓

★ 46.1k▲ 6.6k/yrnpm 125.9k/wk★ +121npm/wk -56.4k

ArenasVector Databases & Memory Stores#2/7

Access

API docs ↗MCP docs ↗

Install

pippip install -U pymilvus

docker

curl -sfL https://raw.githubusercontent.com/milvus-io/milvus/master/scripts/standalone_embed.sh -o standalone_embed.sh && bash standalone_embed.sh start

Compare head-to-head

vs Pinecone vs Weaviate

Alternatives to Milvus →

Try itExperimental

See what an agent can do with Milvus before you ever sign up. Pick a story: recorded sessions replay real probe-harness transcripts; sandboxed self-drive sessions are designed and gated (docs/TRY-IT.md).

$python3 -c 'from pymilvus import MilvusClient; c = MilvusClient("/tmp/pa-milvus-probe.db"); c.create_collection("pa_probe", dimension=8); print("PA_PROBE_OK", c.list_collections())'

recorded session — replayed, not live

▋

recorded 2026-09-04 · exit 0 · captured verbatim by our probe harness, secrets redacted

Verified integrations

Connections to other tracked products — hover a chip for the verbatim evidence quote behind it.

n8nWorkflow Automation

By theme — the product's score on each story themeBy theme

Agenticness — how well agents can access and operate the productAgenticnessevidence →

How well agents can access and operate the product

36.9/100

Automation depth — how much of the product can run unattendedAutomation depthevidence →

How much of the product can run unattended

12.0/100

Data lifecycle — stories about data lifecycle in this arenaData lifecycleevidence →

Stories about data lifecycle in this arena

0.0/100

Deployment modes — stories about deployment modes in this arenaDeployment modesevidence →

Stories about deployment modes in this arena

36.0/100

Embeddings pipeline — stories about embeddings pipeline in this arenaEmbeddings pipelineevidence →

Stories about embeddings pipeline in this arena

30.0/100

Filtering metadata — stories about filtering metadata in this arenaFiltering metadataevidence →

Stories about filtering metadata in this arena

54.0/100

Multi tenancy scale — stories about multi tenancy scale in this arenaMulti tenancy scaleevidence →

Stories about multi tenancy scale in this arena

45.3/100

Openness — open source, data portability, and self-hosting storiesOpennessevidence →

Open source, data portability, and self-hosting stories

64.2/100

Performance latency — stories about performance latency in this arenaPerformance latencyevidence →

Stories about performance latency in this arena

9.0/100

Pricing plans — plan structure and value — what each tier costs and what it unlocksPricing plansevidence →

Plan structure and value — what each tier costs and what it unlocks

26.7/100

Privacy posture — data-handling and privacy storiesPrivacy postureevidence →

Data-handling and privacy stories

12.0/100

Sdk integrations — stories about sdk integrations in this arenaSdk integrationsevidence →

Stories about sdk integrations in this arena

21.0/100

Search quality hybrid — stories about search quality hybrid in this arenaSearch quality hybridevidence →

Stories about search quality hybrid in this arena

59.0/100

Story verdicts — every judged story with its evidenceStory verdicts

What’s free: 2 free · 0 paid · 0 enterprise · 27 not stated in evidence

Sorted by importance (agentic first) (high → low) · 53/53 stories · click a row’s chevron for the rationale and evidence


Connect an agent via an official MCP server G Agent access	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	3	full	8/10	T probed
Drive the product through a documented public API G Agent access	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	3	full	8/10	T probed
Delegate tasks to a built-in AI assistant inside the product G Agentic features	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	3	none	0/10
Plug MCP servers into this product so it can use their tools G Agent access	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	3	n/a	0/10
Build against official SDKs G Agent access	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	full	9/10	X community
Point an agent at llms.txt or agent-oriented docs G Agent access	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	full	9/10	T probed
Run the product headlessly / in CI for automation G Agent access	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	full	7/10	C claimed
Operate the product with natural-language commands G Agentic features	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	partial	6/10	T probed
Issue scoped/least-privilege API credentials for an agent G Agent access	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	partial	5/10	T probed
Download a machine-readable API spec (OpenAPI or equivalent) G Api quality	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	none	0/10
Explore an interactive API reference with runnable examples G Api quality	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	none	0/10
Get AI-generated insights and suggestions from my data inside the product G Agentic features	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	none	0/10
Rely on versioned APIs with a documented deprecation policy G Api quality	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	none	0/10
Set up automations that run autonomously in the background G Agentic features	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	n/a	untested		none yet
Subscribe to events via webhooks G Agent access	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	none	untested		none yet
Use an official CLI G Agent access	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	2	none	untested		none yet
Test against a sandbox environment without touching production data G Api quality	ai-native user	Agenticness — how well agents can access and operate the productAgenticness	1	full	7/10	C claimed
Self-host the core product G	ai-native user	Openness — open source, data portability, and self-hosting storiesOpenness	3	full	9/10	X community
Run approximate nearest-neighbor similarity search over embeddings with configurable distance metrics C Core search	developer	Search quality hybrid — stories about search quality hybrid in this arenaSearch quality hybrid	3	full	8/10	X community
Combine dense vector search with keyword or sparse (BM25-style) signals in one hybrid query with fusion ranking C Hybrid	developer	Search quality hybrid — stories about search quality hybrid in this arenaSearch quality hybrid	3	full	7/10	C claimed
Filter vector search by structured metadata conditions without wrecking recall or latency C Filtering	developer	Filtering metadata — stories about filtering metadata in this arenaFiltering metadata	3	full	7/10	C claimed
Isolate many tenants cheaply using namespaces, partitions, or per-tenant collections with documented limits C Tenancy	platform-engineer	Multi tenancy scale — stories about multi tenancy scale in this arenaMulti tenancy scale	3	partial	6/10	C claimed
Have the database generate embeddings at ingest and query time using built-in or configured model providers, instead of running a separate embedding pipeline C Embeddings	ml-engineer	Embeddings pipeline — stories about embeddings pipeline in this arenaEmbeddings pipeline	3	partial	5/10	C claimed
Export all of my data in open formats and leave G	ai-native user	Openness — open source, data portability, and self-hosting storiesOpenness	3	partial	4/10	C claimed
Define rules that trigger actions automatically on events G	ai-native user	Automation depth — how much of the product can run unattendedAutomation depth	3	none	untested		none yet
Prevent my data from being used to train AI models G	ai-native user	Privacy posture — data-handling and privacy storiesPrivacy posture	3	none	untested		none yet
Run the database embedded in-process or as a lightweight local instance for development and small workloads C Local dev	developer	Deployment modes — stories about deployment modes in this arenaDeployment modes	2	full	9/10	X community
Do everything through the API that I can do in the UI G	ai-native user	Openness — open source, data portability, and self-hosting storiesOpenness	2	full	8/10	T probed
Scale beyond one node with sharding or distributed deployment C Scaling	platform-engineer	Multi tenancy scale — stories about multi tenancy scale in this arenaMulti tenancy scale	2	full	8/10	X community
Enforce granular access control (API keys, roles, per-collection permissions) on database operations C Tenancy	platform-engineer	Multi tenancy scale — stories about multi tenancy scale in this arenaMulti tenancy scale	2	full	7/10	C claimed
Read the product's source under an open license G	ai-native user	Openness — open source, data portability, and self-hosting storiesOpenness	2	fullfree	7/10	T probed
Run keyword/full-text search over documents inside the database without bolting on a separate search engine C Hybrid	developer	Search quality hybrid — stories about search quality hybrid in this arenaSearch quality hybrid	2	full	7/10	C claimed
Perform bulk operations across many items at once G	ai-native user	Automation depth — how much of the product can run unattendedAutomation depth	2	partial	6/10	X community
Choose where my data is stored (region/residency) G	ai-native user	Privacy posture — data-handling and privacy storiesPrivacy posture	2	partial	5/10	C claimed
Express rich filter conditions (ranges, geo, nested boolean logic, array membership) in queries C Filtering	developer	Filtering metadata — stories about filtering metadata in this arenaFiltering metadata	2	partial	5/10	C claimed
Build against official SDKs in the major languages (Python, TypeScript, Go, Java) G Sdks	developer	Sdk integrations — stories about sdk integrations in this arenaSdk integrations	2	partial	4/10	C claimed
Control data retention and deletion G	ai-native user	Privacy posture — data-handling and privacy storiesPrivacy posture	2	partial	4/10	X community
Plug the database into RAG and agent frameworks (LangChain, LlamaIndex, etc.) through maintained first-class integrations C Integrations	ml-engineer	Sdk integrations — stories about sdk integrations in this arenaSdk integrations	2	partial	3/10	T probed
See published benchmarks or measured latency/recall numbers backing the database's performance claims C Benchmarks	platform-engineer	Performance latency — stories about performance latency in this arenaPerformance latency	2	disputed	3/10	D contradicted
Tune index parameters (HNSW graph settings, index types) to trade recall against latency and memory C Index tuning	ml-engineer	Performance latency — stories about performance latency in this arenaPerformance latency	2	partial	3/10	X community
Enable vector quantization or compression to cut memory and storage cost with a documented accuracy trade-off C Index tuning	ml-engineer	Performance latency — stories about performance latency in this arenaPerformance latency	2	none	0/10
Pay serverless usage-based pricing with transparent per-unit costs instead of provisioning fixed clusters G Pricing	developer	Pricing plans — plan structure and value — what each tier costs and what it unlocksPricing plans	2	none	0/10
Replicate data across nodes or zones for high availability with a documented consistency model C Scaling	platform-engineer	Multi tenancy scale — stories about multi tenancy scale in this arenaMulti tenancy scale	2	none	0/10
Upsert and delete records continuously and have changes reflected in search results quickly, with documented freshness/consistency behavior C Freshness	developer	Data lifecycle — stories about data lifecycle in this arenaData lifecycle	2	none	0/10
Use a fully managed cloud version of the database with programmatic provisioning C Managed cloud	developer	Deployment modes — stories about deployment modes in this arenaDeployment modes	2	none	0/10
Back up collections with snapshots and restore them C Backup	platform-engineer	Data lifecycle — stories about data lifecycle in this arenaData lifecycle	2	none	untested		none yet
Bulk-import and bulk-export vectors plus metadata in documented formats C Portability	developer	Data lifecycle — stories about data lifecycle in this arenaData lifecycle	2	none	untested		none yet
Opt out of telemetry and usage tracking G	ai-native user	Privacy posture — data-handling and privacy storiesPrivacy posture	2	none	untested		none yet
Rerank search results with built-in or first-party-integrated reranking models C Reranking	ml-engineer	Search quality hybrid — stories about search quality hybrid in this arenaSearch quality hybrid	2	none	untested		none yet
Schedule recurring jobs or workflows G	ai-native user	Automation depth — how much of the product can run unattendedAutomation depth	2	n/a	untested		none yet
Prototype on a meaningful free tier before paying anything G Pricing	developer	Pricing plans — plan structure and value — what each tier costs and what it unlocksPricing plans	1	fullfree	8/10	C claimed
Deploy to production on Kubernetes with an official Helm chart or operator C Self managed	platform-engineer	Deployment modes — stories about deployment modes in this arenaDeployment modes	1	none	untested		none yet
Version, review, and roll back my automations G	ai-native user	Automation depth — how much of the product can run unattendedAutomation depth	1	none	untested		none yet

Opportunities — the stories that would move this product's scores, from its own judged verdictsOpportunitiestop 8 of 32 stories with headroom

What would move Milvus’s scores — derived from its own judged verdicts, biggest headroom first. Each line quotes what the judge found missing; shipping it (or evidencing it publicly) is the fix.

Agenticness — how well agents can access and operate the productDelegate tasks to a built-in AI assistant inside the product
nonemoves Built-in AIimpact 45
The evidence shows Milvus exposes an MCP server so external AI agents can query it, but this is the reverse of the story — there is no evidence of a built-in AI assistant inside Milvus itself that a user can delegate tasks to.
Automation depth — how much of the product can run unattendedDefine rules that trigger actions automatically on events
nonemoves PA Scoreimpact 30
Missing: any documentation of event-driven triggers, webhook/callback mechanisms, or rule-based automation tied to database events.
Privacy posture — data-handling and privacy storiesPrevent my data from being used to train AI models
nonemoves PA Scoreimpact 30
The evidence pack contains no explicit privacy/data-usage policy addressing whether Milvus or its cloud offering (Zilliz) uses customer data to train AI models.
Agenticness — how well agents can access and operate the productGet AI-generated insights and suggestions from my data inside the product
nonemoves Built-in AIimpact 30
Milvus documentation shows it as a vector search/database engine with MCP-based natural-language query access, but there is no evidence of Milvus itself generating insights, summaries, or suggestions from stored data — it only enables external AI apps to query it, not to produce insights inside the product.
Agenticness — how well agents can access and operate the productUse an official CLI
nonemoves agent-readyimpact 30
Missing: any mention of a dedicated Milvus CLI, its command set, installation, or documentation.
Agenticness — how well agents can access and operate the productSubscribe to events via webhooks
nonemoves agent-readyimpact 30
No evidence anywhere in the pack of Milvus offering webhooks or an event subscription mechanism; it's a vector database with client SDKs, MCP integration, and RBAC, but nothing about outbound event notifications or webhook subscriptions.
Agenticness — how well agents can access and operate the productExplore an interactive API reference with runnable examples
nonemoves API qualityimpact 30
There is no evidence of an interactive API reference (e.g., Swagger/OpenAPI explorer, runnable code sandbox) — the openapi probe explicitly returned 404s at all candidate paths, and documentation consists of static markdown code snippets rather than an interactive, runnable reference.
Agenticness — how well agents can access and operate the productDownload a machine-readable API spec (OpenAPI or equivalent)
nonemoves API qualityimpact 30
The evidence pack shows explicit probes for an OpenAPI/swagger spec on Milvus's site returning 404 for all candidate paths, and no documentation snippet references a downloadable machine-readable API spec (Milvus docs focus on SDK usage, MCP server, RBAC, multi-tenancy, etc.).

Showing the top 8 of 32 — every none/partial verdict in the story verdicts table is headroom.

Think a verdict is wrong? Every verdicts-table row has a Flag link — see the methodology.

Coverage map — which docs area, API section, or community source covers which judged storiesCoverage map4 surfaces · 30 covered stories

Where the cited evidence behind each covered verdict came from — the same citations the verdicts table shows, no extra judging.

GitHub README28 stories

Run the product headlessly / in CI for automation
Connect an agent via an official MCP server
Drive the product through a documented public API
Issue scoped/least-privilege API credentials for an agent
Build against official SDKs
Operate the product with natural-language commands
Test against a sandbox environment without touching production data
Perform bulk operations across many items at once
Run the database embedded in-process or as a lightweight local instance for development and small workloads
Have the database generate embeddings at ingest and query time using built-in or configured model providers, instead of running a separate embedding pipeline
Filter vector search by structured metadata conditions without wrecking recall or latency
Express rich filter conditions (ranges, geo, nested boolean logic, array membership) in queries
Scale beyond one node with sharding or distributed deployment
Enforce granular access control (API keys, roles, per-collection permissions) on database operations
Isolate many tenants cheaply using namespaces, partitions, or per-tenant collections with documented limits
Do everything through the API that I can do in the UI
Export all of my data in open formats and leave
Read the product's source under an open license
Self-host the core product
See published benchmarks or measured latency/recall numbers backing the database's performance claims
Prototype on a meaningful free tier before paying anything
Choose where my data is stored (region/residency)
Control data retention and deletion
Plug the database into RAG and agent frameworks (LangChain, LlamaIndex, etc.) through maintained first-class integrations
Build against official SDKs in the major languages (Python, TypeScript, Go, Java)
Run approximate nearest-neighbor similarity search over embeddings with configurable distance metrics
Run keyword/full-text search over documents inside the database without bolting on a separate search engine
Combine dense vector search with keyword or sparse (BM25-style) signals in one hybrid query with fusion ranking

Hacker News9 stories

Build against official SDKs
Perform bulk operations across many items at once
Run the database embedded in-process or as a lightweight local instance for development and small workloads
Scale beyond one node with sharding or distributed deployment
Self-host the core product
See published benchmarks or measured latency/recall numbers backing the database's performance claims
Tune index parameters (HNSW graph settings, index types) to trade recall against latency and memory
Control data retention and deletion
Run approximate nearest-neighbor similarity search over embeddings with configurable distance metrics

llms.txt2 stories

Point an agent at llms.txt or agent-oriented docs
Read the product's source under an open license

OpenAPI spec2 stories

Drive the product through a documented public API
Do everything through the API that I can do in the UI

Probe proofs — replayable recordings from the probe harnessProbe proofs

Replayable recordings from our probe harness — see the Prove-It protocol to submit one.

$python3 -c 'from pymilvus import MilvusClient; c = MilvusClient("/tmp/pa-milvus-probe.db"); c.create_collection("pa_probe", dimension=8); print("PA_PROBE_OK", c.list_collections())'

reproduced

$ python3 -c 'from pymilvus import MilvusClient; c = MilvusClient("/tmp/pa-milvus-probe.db"); c.create_collection("pa_probe", dimension=8); print("PA_PROBE_OK", c.list_collections())'
PA_PROBE_OK ['pa_probe']

proves: Run the database embedded in-process or as a lightweight local instance for development and small workloads proves: Build against official SDKsrecorded 2026-09-04

Claims vs evidence — vendor claims reconciled against independent verdictsClaims vs evidence

7 of 14 testable claims verified · 2 contradicted → integrity 21/100

16 distinct capability claims found in Milvus’s own claimed-docs/GitHub materials, reconciled against our judge’s independent verdicts.

Verified

Unverified

Contradicted

Undersold

Verified (10)

“Milvus is an open-source vector database suitable for everything from a notebook demo to web-scale search”
Read the product's source under an open license full proof ↗
“Milvus is an open-source vector database suitable for everything from a notebook demo to web-scale search”
Self-host the core product full proof ↗
“Create a local Milvus database instantly by instantiating a MilvusClient pointed at a local file”
Run the database embedded in-process or as a lightweight local instance for development and small workloads full proof ↗
“An MCP server lets AI apps run vector search, manage collections, and retrieve data via natural-language commands”
Connect an agent via an official MCP server full proof ↗
“Same client-side code runs Milvus Lite on a laptop, Standalone in Docker, or Distributed on Kubernetes at billion-scale”
Run the database embedded in-process or as a lightweight local instance for development and small workloads full proof ↗
“Milvus scaled from billion-scale vectors in 2022 to tens of billions in 2023 with consistent stability”
Scale beyond one node with sharding or distributed deployment full proof ↗
“Official pymilvus Python SDK provides a MilvusClient for building applications”
Build against official SDKs full proof ↗
“Milvus Lite shares the same API as Standalone/Distributed and covers CRUD, sparse/dense search, filtering, and hybrid search”
Run the database embedded in-process or as a lightweight local instance for development and small workloads full proof ↗
“Milvus Lite is a lightweight Python library ideal for quick prototyping in notebooks or running on edge devices”
Run the database embedded in-process or as a lightweight local instance for development and small workloads full proof ↗
“Client search API accepts a batch of query vectors and a result limit for ANN search”
Run approximate nearest-neighbor similarity search over embeddings with configurable distance metrics full proof ↗

Unverified (5)

“Accepts raw text input and automatically converts it to sparse embeddings without manual embedding generation”
Have the database generate embeddings at ingest and query time using built-in or configured model providers, instead of running a separate embedding pipeline partial proof ↗
“Supports searching across multiple vector fields by running several ANN searches simultaneously”
Combine dense vector search with keyword or sparse (BM25-style) signals in one hybrid query with fusion ranking full proof ↗
“Search requests can include metadata filters applied before ANN search to shrink the search scope”
Filter vector search by structured metadata conditions without wrecking recall or latency full proof ↗
“Offers four multi-tenancy strategies with different scalability/isolation/flexibility trade-offs”
Isolate many tenants cheaply using namespaces, partitions, or per-tenant collections with documented limits partial proof ↗
“RBAC lets admins finely control user operations at collection, database, and instance levels”
Enforce granular access control (API keys, roles, per-collection permissions) on database operations full proof ↗

Contradicted (2)

“Same client-side code runs Milvus Lite on a laptop, Standalone in Docker, or Distributed on Kubernetes at billion-scale”
Deploy to production on Kubernetes with an official Helm chart or operator none
“Milvus delivers 30%-70% better performance than FAISS and HNSWLib per vendor benchmarks”
See published benchmarks or measured latency/recall numbers backing the database's performance claims disputed proof ↗

Undersold (17)

Point an agent at llms.txt or agent-oriented docs full proof ↗
Run the product headlessly / in CI for automation full proof ↗
Drive the product through a documented public API full proof ↗
Issue scoped/least-privilege API credentials for an agent partial proof ↗
Operate the product with natural-language commands partial proof ↗
Test against a sandbox environment without touching production data full proof ↗
Perform bulk operations across many items at once partial proof ↗
Express rich filter conditions (ranges, geo, nested boolean logic, array membership) in queries partial proof ↗
Do everything through the API that I can do in the UI full proof ↗
Export all of my data in open formats and leave partial proof ↗
Tune index parameters (HNSW graph settings, index types) to trade recall against latency and memory partial proof ↗
Prototype on a meaningful free tier before paying anything full proof ↗
Choose where my data is stored (region/residency)partial proof ↗
Control data retention and deletion partial proof ↗
Plug the database into RAG and agent frameworks (LangChain, LlamaIndex, etc.) through maintained first-class integrations partial proof ↗
Build against official SDKs in the major languages (Python, TypeScript, Go, Java)partial proof ↗
Run keyword/full-text search over documents inside the database without bolting on a separate search engine full proof ↗

Claims outside our story set (1)

Real capability claims found in Milvus’s own materials, but no story in this arena’s taxonomy covers them yet — that’s feedback on the taxonomy, not a mark against the product.

“Supports a Standalone deployment mode for single-machine setups”
source ↗

Suggest a story for these →

Pricing signals

$63per 1M vectors/monthpay-as-you-goPerformance-optimized dedicated cluster ratesource ↗as of 2026-09-07
$16per 1M vectors/monthpay-as-you-goCapacity-optimized dedicated cluster ratesource ↗as of 2026-09-07
$5per 1M vectors/monthpay-as-you-goTiered-storage dedicated cluster ratesource ↗as of 2026-09-07
$126per GB-monthpay-as-you-goStandard plan dedicated rate per GB of storagesource ↗as of 2026-09-07
$197per month (entry plan)entry planEnterprise plan starting dedicated pricesource ↗as of 2026-09-07
freeper GB-monthfree tierFree tier includes 5 GB storage and 2.5M vCUs per month at no costsource ↗as of 2026-09-07

Extracted verbatim from the vendor’s own pricing page — hover a figure for the exact quote.

Business model

open-sourcefree-tierusage-basedenterprise-custom

Apache-2.0 open-source vector database under LF AI & Data, free to self-host from laptop (Milvus Lite) to cluster; Zilliz Cloud sells the managed version with free, serverless, and dedicated tiers.

pricing ↗

Score trend

How this product’s scores have moved as evidence and verdicts are re-derived — a point per change, not per day.

PA Score36 (Sep 4 '26)36 (Sep 15 '26)

Agent-ready58 (Sep 4 '26)58 (Sep 15 '26)

Try Experimental

Run it in the microterminal →

Recorded agent sessions — and a live MCP handshake where the vendor ships one.

Flag

⚑ Flag a verdict

Think a verdict is wrong? Opens a prefilled GitHub issue — or use the ⚑ next to any verdict above.

Badge

Embed this product's score badge →

Hotlinked SVG — always shows the live current score.

For agents

This page as markdown /llms.txt

Data

Evidence (JSON)Verdicts (JSON)

Agent surface uptime llms.txt 100% (30d, checked every 6h since Sep 8 '26)

Milvus

Try itExperimental

Verified integrations

By theme — the product's score on each story themeBy theme

Story verdicts — every judged story with its evidenceStory verdicts

🤖Agenticness — how well agents can access and operate the productAgenticness

⚙️Automation depth — how much of the product can run unattendedAutomation depth

🗄️Data lifecycle — stories about data lifecycle in this arenaData lifecycle

🚀Deployment modes — stories about deployment modes in this arenaDeployment modes

✨Embeddings pipeline — stories about embeddings pipeline in this arenaEmbeddings pipeline

🗄️Filtering metadata — stories about filtering metadata in this arenaFiltering metadata

🔑Multi tenancy scale — stories about multi tenancy scale in this arenaMulti tenancy scale

🔓Openness — open source, data portability, and self-hosting storiesOpenness

⚡Performance latency — stories about performance latency in this arenaPerformance latency

💲Pricing plans — plan structure and value — what each tier costs and what it unlocksPricing plans

🔒Privacy posture — data-handling and privacy storiesPrivacy posture

🔌Sdk integrations — stories about sdk integrations in this arenaSdk integrations

🔎Search quality hybrid — stories about search quality hybrid in this arenaSearch quality hybrid

Opportunities — the stories that would move this product's scores, from its own judged verdictsOpportunitiestop 8 of 32 stories with headroom

Coverage map — which docs area, API section, or community source covers which judged storiesCoverage map4 surfaces · 30 covered stories

Probe proofs — replayable recordings from the probe harnessProbe proofs

Claims vs evidence — vendor claims reconciled against independent verdictsClaims vs evidence

Pricing signals

Business model

Score trend

Agenticness — how well agents can access and operate the productAgenticness

Automation depth — how much of the product can run unattendedAutomation depth

Data lifecycle — stories about data lifecycle in this arenaData lifecycle

Deployment modes — stories about deployment modes in this arenaDeployment modes

Embeddings pipeline — stories about embeddings pipeline in this arenaEmbeddings pipeline

Filtering metadata — stories about filtering metadata in this arenaFiltering metadata

Multi tenancy scale — stories about multi tenancy scale in this arenaMulti tenancy scale

Openness — open source, data portability, and self-hosting storiesOpenness

Performance latency — stories about performance latency in this arenaPerformance latency

Pricing plans — plan structure and value — what each tier costs and what it unlocksPricing plans

Privacy posture — data-handling and privacy storiesPrivacy posture

Sdk integrations — stories about sdk integrations in this arenaSdk integrations

Search quality hybrid — stories about search quality hybrid in this arenaSearch quality hybrid