Skip to content
ProductArena

Rank #1 of 7 in Vector Databases & Memory Stores

Chroma logo

Chroma

Open Source

Chroma (Trychroma, Inc.)

trychroma.com ↗Try it →
PA Score41/100AGENT-READY55/100BUILT-IN AI10/100API38/100
MCP✓CLI✓API✓
★ 29.3k▲ 7.4k/yrnpm 205.3k/wkpypi 1.5M/wk★ +66npm/wk -76.6kpypi/wk -84.2k
ArenasVector Databases & Memory Stores#1/7

Access

Open app ↗API docs ↗CLI docs ↗MCP docs ↗

Install

pippip install chromadb
npmnpm install chromadb

Compare head-to-head

Pinecone logovs Pineconevs WeaviateQdrant logovs QdrantMilvus logovs MilvusHelixDB logovs HelixDBLanceDB logovs LanceDB

Alternatives to Chroma →

Showcase

Chroma homepage screenshot
homepage · captured Sep 2026 · view live ↗
Chroma docs screenshot
docs · captured Sep 2026 · view live ↗

Try itExperimental

See what an agent can do with Chroma 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).

$chroma --helprecorded 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.

Claude logoClaudeAI Assistants

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

41.3/100

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

How much of the product can run unattended

24.0/100

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

Stories about data lifecycle in this arena

6.0/100

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

Stories about deployment modes in this arena

46.4/100

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

Stories about embeddings pipeline in this arena

80.0/100

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

Stories about filtering metadata in this arena

21.6/100

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

Stories about multi tenancy scale in this arena

22.0/100

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

Open source, data portability, and self-hosting stories

60.0/100

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

Stories about performance latency in this arena

0.0/100

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

73.3/100

Privacy posture — data-handling and privacy storiesPrivacy postureevidence →

Data-handling and privacy stories

13.3/100

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

Stories about sdk integrations in this arena

27.0/100

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

Stories about search quality hybrid in this arena

23.4/100

Story verdicts — every judged story with its evidenceStory verdicts

What’s free: 6 free · 5 paid · 0 enterprise · 22 not stated in evidence

?

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

Drive the product through a documented public API G

Agent access

ai-native userAgenticness — how well agents can access and operate the productAgenticness3full9/10T

Connect an agent via an official MCP server G

Agent access

ai-native userAgenticness — how well agents can access and operate the productAgenticness3full8/10T

Plug MCP servers into this product so it can use their tools G

Agent access

ai-native userAgenticness — how well agents can access and operate the productAgenticness3n/a0/10

Delegate tasks to a built-in AI assistant inside the product G

Agentic features

ai-native userAgenticness — how well agents can access and operate the productAgenticness3noneuntestednone yet

Download a machine-readable API spec (OpenAPI or equivalent) G

Api quality

ai-native userAgenticness — how well agents can access and operate the productAgenticness2full9/10T

Point an agent at llms.txt or agent-oriented docs G

Agent access

ai-native userAgenticness — how well agents can access and operate the productAgenticness2full8/10T

Use an official CLI G

Agent access

ai-native userAgenticness — how well agents can access and operate the productAgenticness2full8/10T

Build against official SDKs G

Agent access

ai-native userAgenticness — how well agents can access and operate the productAgenticness2partial7/10T

Operate the product with natural-language commands G

Agentic features

ai-native userAgenticness — how well agents can access and operate the productAgenticness2partial6/10T

Run the product headlessly / in CI for automation G

Agent access

ai-native userAgenticness — how well agents can access and operate the productAgenticness2partialfree6/10T

Explore an interactive API reference with runnable examples G

Api quality

ai-native userAgenticness — how well agents can access and operate the productAgenticness2partial4/10T

Get AI-generated insights and suggestions from my data inside the product G

Agentic features

ai-native userAgenticness — how well agents can access and operate the productAgenticness2noneuntestednone yet

Issue scoped/least-privilege API credentials for an agent G

Agent access

ai-native userAgenticness — how well agents can access and operate the productAgenticness2noneuntestednone yet

Rely on versioned APIs with a documented deprecation policy G

Api quality

ai-native userAgenticness — how well agents can access and operate the productAgenticness2noneuntestednone yet

Set up automations that run autonomously in the background G

Agentic features

ai-native userAgenticness — how well agents can access and operate the productAgenticness2n/auntestednone yet

Subscribe to events via webhooks G

Agent access

ai-native userAgenticness — how well agents can access and operate the productAgenticness2noneuntestednone yet

Test against a sandbox environment without touching production data G

Api quality

ai-native userAgenticness — how well agents can access and operate the productAgenticness1partialpaid6/10X

Self-host the core product G

ai-native userOpenness — open source, data portability, and self-hosting storiesOpenness3fullfree9/10T

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-engineerEmbeddings pipeline — stories about embeddings pipeline in this arenaEmbeddings pipeline3full8/10C

Export all of my data in open formats and leave G

ai-native userOpenness — open source, data portability, and self-hosting storiesOpenness3partialfree5/10X

Isolate many tenants cheaply using namespaces, partitions, or per-tenant collections with documented limits C

Tenancy

platform-engineerMulti tenancy scale — stories about multi tenancy scale in this arenaMulti tenancy scale3partialpaid5/10X

Run approximate nearest-neighbor similarity search over embeddings with configurable distance metrics C

Core search

developerSearch quality hybrid — stories about search quality hybrid in this arenaSearch quality hybrid3partial5/10C

Combine dense vector search with keyword or sparse (BM25-style) signals in one hybrid query with fusion ranking C

Hybrid

developerSearch quality hybrid — stories about search quality hybrid in this arenaSearch quality hybrid3partial4/10C

Filter vector search by structured metadata conditions without wrecking recall or latency C

Filtering

developerFiltering metadata — stories about filtering metadata in this arenaFiltering metadata3partial4/10C

Define rules that trigger actions automatically on events G

ai-native userAutomation depth — how much of the product can run unattendedAutomation depth3n/auntestednone yet

Prevent my data from being used to train AI models G

ai-native userPrivacy posture — data-handling and privacy storiesPrivacy posture3noneuntestednone yet

Read the product's source under an open license G

ai-native userOpenness — open source, data portability, and self-hosting storiesOpenness2fullfree9/10X

Run the database embedded in-process or as a lightweight local instance for development and small workloads C

Local dev

developerDeployment modes — stories about deployment modes in this arenaDeployment modes2fullfree8/10T

Pay serverless usage-based pricing with transparent per-unit costs instead of provisioning fixed clusters G

Pricing

developerPricing plans — plan structure and value — what each tier costs and what it unlocksPricing plans2fullpaid7/10X

Choose where my data is stored (region/residency) G

ai-native userPrivacy posture — data-handling and privacy storiesPrivacy posture2partial6/10C

Perform bulk operations across many items at once G

ai-native userAutomation depth — how much of the product can run unattendedAutomation depth2partial6/10C

Plug the database into RAG and agent frameworks (LangChain, LlamaIndex, etc.) through maintained first-class integrations C

Integrations

ml-engineerSdk integrations — stories about sdk integrations in this arenaSdk integrations2partial6/10T

Run keyword/full-text search over documents inside the database without bolting on a separate search engine C

Hybrid

developerSearch quality hybrid — stories about search quality hybrid in this arenaSearch quality hybrid2partial6/10X

Scale beyond one node with sharding or distributed deployment C

Scaling

platform-engineerMulti tenancy scale — stories about multi tenancy scale in this arenaMulti tenancy scale2partial6/10X

Use a fully managed cloud version of the database with programmatic provisioning C

Managed cloud

developerDeployment modes — stories about deployment modes in this arenaDeployment modes2partialpaid6/10X

Do everything through the API that I can do in the UI G

ai-native userOpenness — open source, data portability, and self-hosting storiesOpenness2partial5/10T

Control data retention and deletion G

ai-native userPrivacy posture — data-handling and privacy storiesPrivacy posture2partial4/10C

Back up collections with snapshots and restore them C

Backup

platform-engineerData lifecycle — stories about data lifecycle in this arenaData lifecycle2partialpaid3/10C

Build against official SDKs in the major languages (Python, TypeScript, Go, Java) G

Sdks

developerSdk integrations — stories about sdk integrations in this arenaSdk integrations2partial3/10C

Express rich filter conditions (ranges, geo, nested boolean logic, array membership) in queries C

Filtering

developerFiltering metadata — stories about filtering metadata in this arenaFiltering metadata2partial3/10C

Replicate data across nodes or zones for high availability with a documented consistency model C

Scaling

platform-engineerMulti tenancy scale — stories about multi tenancy scale in this arenaMulti tenancy scale2partial3/10X

Enable vector quantization or compression to cut memory and storage cost with a documented accuracy trade-off C

Index tuning

ml-engineerPerformance latency — stories about performance latency in this arenaPerformance latency2none0/10

Rerank search results with built-in or first-party-integrated reranking models C

Reranking

ml-engineerSearch quality hybrid — stories about search quality hybrid in this arenaSearch quality hybrid2none0/10

Bulk-import and bulk-export vectors plus metadata in documented formats C

Portability

developerData lifecycle — stories about data lifecycle in this arenaData lifecycle2noneuntestednone yet

Enforce granular access control (API keys, roles, per-collection permissions) on database operations C

Tenancy

platform-engineerMulti tenancy scale — stories about multi tenancy scale in this arenaMulti tenancy scale2noneuntestednone yet

Opt out of telemetry and usage tracking G

ai-native userPrivacy posture — data-handling and privacy storiesPrivacy posture2noneuntestednone yet

Schedule recurring jobs or workflows G

ai-native userAutomation depth — how much of the product can run unattendedAutomation depth2n/auntestednone yet

See published benchmarks or measured latency/recall numbers backing the database's performance claims C

Benchmarks

platform-engineerPerformance latency — stories about performance latency in this arenaPerformance latency2noneuntestednone yet

Tune index parameters (HNSW graph settings, index types) to trade recall against latency and memory C

Index tuning

ml-engineerPerformance latency — stories about performance latency in this arenaPerformance latency2noneuntestednone yet

Upsert and delete records continuously and have changes reflected in search results quickly, with documented freshness/consistency behavior C

Freshness

developerData lifecycle — stories about data lifecycle in this arenaData lifecycle2noneuntestednone yet

Prototype on a meaningful free tier before paying anything G

Pricing

developerPricing plans — plan structure and value — what each tier costs and what it unlocksPricing plans1fullfree8/10X

Deploy to production on Kubernetes with an official Helm chart or operator C

Self managed

platform-engineerDeployment modes — stories about deployment modes in this arenaDeployment modes1none0/10

Version, review, and roll back my automations G

ai-native userAutomation depth — how much of the product can run unattendedAutomation depth1noneuntestednone yet

Opportunities — the stories that would move this product's scores, from its own judged verdictsOpportunitiestop 8 of 38 stories with headroom

What would move Chroma’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.

  1. 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 axis applies to this product kind (peer products hold positive or none verdicts on this story), so lack of evidence for an applicable capability is "none", never "na".

  2. Privacy posture — data-handling and privacy storiesPrevent my data from being used to train AI models

    nonemoves PA Scoreimpact 30

    Chroma is a vector database; the evidence pack contains no statement about data-training opt-out policies, data usage terms, or privacy commitments regarding whether user data is used to train AI models.

  3. 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

    Chroma is positioned as a vector/embedding database and retrieval backend (storage, indexing, querying, MCP-based agent access) rather than a product that itself surfaces AI-generated insights or suggestions to the user; none of the evidence describes built-in analytics, summarization, or recommendation features inside Chroma's own interface.

  4. Agenticness — how well agents can access and operate the productIssue scoped/least-privilege API credentials for an agent

    nonemoves agent-readyimpact 30

    No evidence of scoped or least-privilege API credential issuance for agents; the evidence pack covers embeddings, querying, CLI, MCP server, and pricing but never mentions API key scoping, RBAC, or per-agent credential management in Chroma Cloud or self-hosted deployments.

  5. Agenticness — how well agents can access and operate the productSubscribe to events via webhooks

    nonemoves agent-readyimpact 30

    No evidence in the pack mentions webhooks or event subscription mechanisms for Chroma; the product is a vector database and its documented integrations (MCP, CLI, APIs) do not include any webhook/event-push feature.

  6. Agenticness — how well agents can access and operate the productRely on versioned APIs with a documented deprecation policy

    nonemoves API qualityimpact 30

    Missing: any mention of API versioning, changelog, or deprecation/backward-compatibility policy.

  7. Multi tenancy scale — stories about multi tenancy scale in this arenaEnforce granular access control (API keys, roles, per-collection permissions) on database operations

    nonemoves PA Scoreimpact 20

    No evidence in the pack mentions API keys, roles, RBAC, or per-collection permission scoping; docs cover embedding, querying, CLI, MCP, forking, and cloud pricing but nothing about access control mechanisms.

  8. Data lifecycle — stories about data lifecycle in this arenaBulk-import and bulk-export vectors plus metadata in documented formats

    nonemoves PA Scoreimpact 20

    No evidence pack item documents a bulk-import or bulk-export feature, file format spec, or CLI/API command for moving vectors+metadata in/out of Chroma; forking (chroma-docs-8/18/26) is copy-on-write cloning, not data export/import.

Showing the top 8 of 38 — 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 map9 surfaces · 33 covered stories

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

docs20 stories

docs.trychroma.com16 stories

Hacker News12 stories

trychroma.com10 stories

Cloud docs9 stories

OpenAPI spec5 stories

llms.txt4 stories

GitHub README4 stories

Integrations docs3 stories

Probe proofs — replayable recordings from the probe harnessProbe proofs

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

$chroma --helpreproduced
$ chroma --help
A CLI for Chroma

Usage: chroma <COMMAND>

Commands:
  browse   Browse Chroma collections
  copy     Copy collection between local and Chroma Cloud
  db       Manage Chroma Cloud databases
  docs     Open Chroma online documentation
  install  Install sample applications
  login    Log in to Chroma Cloud
  profile  Manage Chroma Cloud profiles
  run      Start a local Chroma server
  support  Open the Chroma Discord
  update   Check for Chroma CLI updates
  vacuum   Vacuum a local Chroma persistent directory
  help     Print this message or the help of the given subcommand(s)

Options:
  -h, --help     Print help
  -V, --version  Print version
proves: Use an official CLIrecorded 2026-09-04
$chroma run --path /tmp/pa-chroma-probe --port 8765 & curl -X POST localhost:8765/api/v2/tenants/default_tenant/databases/default_database/collections -d '{"name":"pa_probe"}' && curl localhost:8765/api/v2/tenants/default_tenant/databases/default_database/collectionsreproduced
$ chroma run --path /tmp/pa-chroma-probe --port 8765 & curl -X POST localhost:8765/api/v2/tenants/default_tenant/databases/default_database/collections -d '{"name":"pa_probe"}' && curl localhost:8765/api/v2/tenants/default_tenant/databases/default_database/collections
{"nanosecond heartbeat":1788560721960487000}
{"id":"cb4da00a-c4cb-4e9d-9781-9b5621b9aaec","name":"pa_probe","configuration_json":{"hnsw":{"space":"l2","ef_construction":100,"ef_search":100,"max_neighbors":16,"resize_factor":1.2,"sync_threshold":1000},"spann":null,"embedding_function":
[{"id":"cb4da00a-c4cb-4e9d-9781-9b5621b9aaec","name":"pa_probe","configuration_json":{"hnsw":{"space":"l2","ef_construction":100,"ef_search":100,"max_neighbors":16,"resize_factor":1.2,"sync_threshold":1000},"spann":null,"embedding_function"
proves: Run the database embedded in-process or as a lightweight local instance for development and small workloadsproves: Drive the product through a documented public APIrecorded 2026-09-04
$echo '<jsonrpc initialize>' | uvx chroma-mcp --client-type ephemeralreproduced
$ echo '<jsonrpc initialize>' | uvx chroma-mcp --client-type ephemeral
⠋ Resolving dependencies...                                                     
⠙ Resolving dependencies...                                                     
⠋ Resolving dependencies...                                                     
⠙ Resolving dependencies...                                                     
⠙ chroma-mcp==0.2.6                                                             
⠙ mcp==1.6.0                                                                    
⠙ mcp==1.6.0                                                                    
⠙ chromadb==1.5.9                                                               
⠙ cohere==7.1.1                                                                 
⠙ httpx==0.28.1                                                                 
⠙ openai==3.8.0                                                                 
⠙ pillow==12.3.0                                                                
⠙ python-dotenv==1.2.3                                                          
⠙ typing-extensions==4.16.0                                                     
⠙ voyageai==0.5.0                                                               
⠙ anyio==4.15.0                                                                 
⠙ typing-extensions==4.16.0                                                     
⠙ httpx-sse==0.4.3                                                              
⠙ pydantic-settings==2.15.0                                                     
⠙ pydantic==2.13.5                                                              
⠙ pydantic-core==2.46.5                                                         
⠙ sse-starlette==3.4.10                                                         
/Users/judegomila/.cache/uv/archive-v0/JD11Ic7cM6QK165B/lib/python3.13/site-packages/pydantic_settings/sources/utils.py:47: IncompleteFieldDefinitionWarning: Field 'lifespan' has an incomplete definition: its annotation contains an unresolved forward reference, so settings sources may fail to correctly resolve its value. Call `model_rebuild()` on the model where the field is defined, once all the referenced types are defined.
  warnings.warn(
Successfully initialized Chroma client
Starting MCP server
{"jsonrpc":"2.0","id":1,"result":{"protocolVersion":"2024-11-05","capabilities":{"experimental":{},"prompts":{"listChanged":false},"resources":{"subscribe":false,"listChanged":false},"tools":{"listChanged":false}},"serverInfo":{"name":"chroma","version":"1.6.0"}}}
proves: Connect an agent via an official MCP serverrecorded 2026-09-04

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

10 of 16 testable claims verified · 0 contradicted → integrity 63/100

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

10

Verified

6

Unverified

0

Contradicted

17

Undersold

Verified (11)
Unverified (7)
Undersold (17)
Claims outside our story set (3)

Real capability claims found in Chroma’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 collection forking for dataset versioning, A/B testing, and roll-outs”

    source ↗

  • “Forks are copy-on-write, only charging for incremental storage written after the fork”

    source ↗

  • “Can index and search images, audio, and other modalities alongside text”

    source ↗
Suggest a story for these →

Pricing signals

  • $2.5per GiB writtenpay-as-you-goCost per logical GiB written via add/update/upsertsource ↗as of 2026-09-07
  • $0.33per GB-monthpay-as-you-goStorage cost per GiB per month, prorated hourlysource ↗as of 2026-09-07
  • $0.0075per TB queriedpay-as-you-goRead cost per TiB of data queriedsource ↗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-based

Apache-2.0 open-source embedding database that runs embedded or as a server for free; Chroma Cloud charges usage-based pricing for writes, queries, and storage with $5 starting credit.

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 Score44 (Sep 4 '26)41 (Sep 15 '26)
Agent-ready55 (Sep 4 '26)55 (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% · openapi.json 100% (30d, checked every 6h since Sep 8 '26)