Head to head · Agent tracing · October 2026 research run

Arize Phoenix vs Laminar API + MCP

Arize Phoenix scores 75.8 (BB) on agent readiness against Laminar API + MCP's 58.9 (C), and leads in 6 of 7 scored categories. Both do agent tracing.

Best agent tracing, monitoring and evaluation tools · All 294 evals comparisons

Which one, for what

Arize Phoenix BB

Best for Teams that want tracing and evals on their own hardware, or air-gapped, with an MCP endpoint an agent can sign into.

Ahead on

  • Reliability, 73 against 52
  • Agent ergonomics, 90 against 80
  • Security & auth, 58 against 44
  • Payments & pricing, 90 against 45
  • Transparency & trust, 75 against 64

Also in its favour

  • Agent-ready, a grade of BB or better
  • Runs on your own machine

Watch for

Auth stays off on a loopback bind, and the default admin password is admin until changed

Laminar API + MCP C

Best for Teams running browser agents who want session replay next to traces, and for agents that read traces through SQL.

Also in its favour

  • A hosted endpoint, with nothing to install

Watch for

A cross-tenant SQL execution and export hole was fixed on 27 August 2026 in a commit, with no advisory

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.

Payments & pricing Arize Phoenix +45

Arize Phoenix 90

No x402 (0). Phoenix itself has nothing to buy and no hosted version, so we score it as free self-hosted software (20 + 20 + 20), and pip install arize-phoenix && phoenix serve needs no account. Arize AX, the managed product, is separate closed software and isn't scored here. That's a judgement call. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 60 is 90.

Laminar API + MCP 45

No x402 (0). Per-GB prices and Signals token prices published without login (20). The Free plan exists with 1 GB and 7 days, but the pricing page doesn't say whether a card is needed, so half (10). A person signs up in a browser to create a project key (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 30 is 45.

Reliability Arize Phoenix +21

Arize Phoenix 73

Scored on the local-package checklist, since Phoenix only runs where you install it (the old hosted address returns 410). Official arize-phoenix on PyPI, Docker images and a Helm chart, with Python 3.11 to 3.15 stated in pyproject.toml on 1 November (20). Python, TypeScript, Playwright and Helm CI run in public. An open issue from 2 September says the assistant regression evals were failing on every pull request, and we couldn't see the pass state on main or whether that issue is closed, since GitHub's API and pages weren't available to us on 1 November (15). 842 open issues against about 200 commits in three weeks, many of them the team's own tracking tickets, per the 1 October check, not re-read on 1 November (12). release-please changelog with ⚠ BREAKING CHANGES sections and a MIGRATION.md. 21.0.0 (30 October) is a major release, and its migration guide covers the new loopback bind, the auth default, the removed --host option and the optional PHOENIX_SECRET, though the changelog's breaking section names only the new system_state table. 20.13.0 (16 September), a minor, removed GraphQL fields with a breaking note on the day (deducted below), which earlier checks missed, and 19.4.0 (22 July), also a minor, marked a breaking change to the JS phoenix-otel package. Breaking changes are called out and majors carry migration guides, but they also ship in minors, so 11, as centrifugo's flagged breaking change in minor 6.9.0 was read (11). Version 21.0.0 (15). 20+15+12+11+15 = 73.

Laminar API + MCP 52

Laminar's terms (last updated 11 August 2026) bar any spider, robot or scraper that accesses the Services other than its documented APIs, SDKs and CLI, so status.laminar.sh was not re-read on 4 November and the 1 October readings are kept. Better Stack page whose only components are the homepage and an API echo probe, with nothing for ingestion, SQL or MCP, kept at 12 as partly met. Per the 1 October check the window from 6 August to 1 October shows only one planned 30-minute Postgres upgrade on 8 August, and no unread stretch is credited beyond it (20). The SQL docs still say queries are rate limited per project with no numbers (0). The SQL API documents 429 as 'Project rate limit exceeded. Retry with backoff', which is backoff guidance alone. Annotation queue items take an idempotencyKey, but no retry guidance covers ingestion (10). No SLA found (0). API, SQL and MCP are GA (10). 12+20+0+10+0+10 = 52.

Security & auth Arize Phoenix +14

Arize Phoenix 58

Since 21.0.0 (30 October) Phoenix binds to 127.0.0.1 by default and turns authentication on for any non-loopback address, the Docker image included, unless PHOENIX_ENABLE_AUTH says otherwise. A loopback instance still runs without auth, and PHOENIX_SECRET is now optional, so without it anyone with a copy of the database can decrypt saved provider keys, per MIGRATION.md. With auth on, the MCP endpoint uses Phoenix's own OAuth 2.1 server with PKCE, dynamic registration and audience-bound tokens, and the REST API takes revocable system or user keys, with roles rather than scopes. The default admin password is still admin, and since 21.0.0 it must be changed before a session opens (20). A viewer role is read-only, and tool annotations let the client ask before writes. executeGraphqlMutation (20.20.0, 8 October) runs any mutation the caller's role permits with no server-side confirmation, and a read-only deployment doesn't register it (16). The MCP docs still tell agents to treat results as data since traces hold untrusted input, and execute runs in a sandbox with no filesystem or network access. Guidance only, with approval left to the client (10). No audit log found. 20.23.0 can export server logs over OTLP and 20.27.0 redacts personal data from them, which isn't a per-call audit trail (0). SECURITY.md with a disclosure address. Arize's bug bounty excludes the open-source repositories. Fixes go out in the changelog rather than as advisories (21.0.0 signs and binds the OAuth2 login-context cookie, 20.27.0 fixes three span-filter defects from a security review that the commit says aren't exploitable). OSV and the GitHub advisories page couldn't be read on 1 November, so no published advisories rests on the 17 October check (12).

Laminar API + MCP 44

One project API key as a Bearer token, revocable, scoped to one project, with no read-only key type (20). MCP still has three tools and no write tools, and SQL is SELECT-only on a read-only database client (16). Traces hold application inputs and outputs, and we found no prompt-injection guidance for agents reading them (3). No audit log found (0). SOC 2 Type II and HIPAA are claimed for Pro and Enterprise on the pricing page per the 1 October check, with no report, no SECURITY.md in any of the three repositories, no security.txt per the 1 October check and no advisories in OSV for lmnr or Laminar. A bare SOC 2 claim with no disclosure route takes 5 in this run (5). 20+16+3+0+5 = 44.

Transparency & trust Arize Phoenix +11

Arize Phoenix 75

Elastic License 2.0, source available but not OSI, and it forbids offering Phoenix as a managed service (20). Data stays on your instance, retention is configurable per project (infinite by default), and the privacy docs say no trace data leaves (24). Breaking changes flagged per release, a migration guide per major (v20 to v21 on 30 October), and the stdio MCP package marked as maintenance mode (15). Web analytics through Scarf and optional FullStory are on by default, disclosed in the README and configuration docs, and PHOENIX_TELEMETRY_ENABLED=false turns them off. Since 21.0.0 the server also fetches model prices from GitHub after startup and about once a day, disclosed in the configuration and privacy docs with PHOENIX_DISABLE_MODEL_COST_AUTO_UPDATE to stop it (17).

Laminar API + MCP 64

Apache-2.0, with Signals and Slack alerts held back from the open-source images behind a licence key (28). The privacy policy covers account data in general terms with no retention period for trace data, no subprocessor list and no DPA link we could find. Retention per plan is on the pricing page and PII redaction is documented (12). No deprecation policy. The move from lmnr.ai to laminar.sh while the API stayed on api.lmnr.ai wasn't dated in the changelog (5). Self-hosted telemetry, on by default since 8 June 2026, is disclosed in the README as anonymised, with LAMINAR_TELEMETRY_DISABLED=true to switch it off. The code sends a per-deployment UUID, version, environment and row counts, with the source IP dropped. Until v0.3.2 (16 October) the documented opt-out didn't reach the Compose frontend containers (7280cc17). No subprocessor or data-location list for the cloud (12).

Agent ergonomics Arize Phoenix +10

Arize Phoenix 90

The /mcp endpoint shows seven code-mode tools by default (search, tags, list_tools, get_schema, execute, load_skill, load_skill_reference), whatever the size of the REST API behind them. Since 20.20.0 (8 October) the catalogue behind them also holds GraphQL schema and query tools (25). Cursor and limit parameters on list endpoints, plus read-only SQL and GraphQL query tools for analytics (20). SQL errors come back with teaching hints. REST errors are plain FastAPI details (14). Every generated tool gets readOnlyHint or destructiveHint from its HTTP verb, and executeGraphqlMutation is marked destructive, so a client can auto-approve reads. execute can reach writes, which the annotations can't separate (16). Few required parameters. Python and TypeScript clients and a CLI (15).

Laminar API + MCP 80

Three MCP tools, but the SQL tool's description embeds the whole table schema, so the list costs more context than the count suggests (18). Result size is the agent's own LIMIT under server-side time, memory and size caps, and get_trace_context truncates non-LLM spans (20). Tool failures come back with isError and a message. HTTP errors are a single error field (17). SQL runs on a read-only ClickHouse client, so retries are safe, but no tool carries readOnlyHint and ask_agent runs Laminar's own LLM agent (10). One required argument per tool. Python and TypeScript SDKs and a CLI (15).

Score by category

CategoryWeightArize PhoenixLaminar API + MCPEdge
Reliability16%7352Arize Phoenix +21
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%8888even
Agent ergonomics13%9080Arize Phoenix +10
Security & auth14%5844Arize Phoenix +14
Payments & pricing10%9045Arize Phoenix +45
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8483Arize Phoenix +1
Transparency & trust7%7564Arize Phoenix +11
Negative events≤15-3-5
TotalBB 75.8/100C 58.9/100

Facts side by side

FactArize PhoenixLaminar API + MCP
KindHTTP APIHTTP API
VendorArize AILaminar (LMNR AI)
Hosted endpointno (local only)https://api.lmnr.ai/v1
TransportsHTTP, Streamable HTTP, stdioHTTP, Streamable HTTP
AuthOAuth or keyAPI key
PricingFreeFreemium
x402nono
LicenceElastic-2.0Apache-2.0
Tools exposed53
Read-only variant documentednoyes
llms.txtyesyes
Last release2026-10-302026-11-03
Terms last updatedno date given2026-08-11
Privacy policy last updated2026-08-112026-08-11
Customer content may train modelsnot found in the textyes, with an opt-out
Terms restrict automated accessyesyes
Terms restrict benchmarkingyesyes
Terms or service can change without noticenot found in the textyes
Arbitration or class-action waiveryesnot found in the text
Popularity12k stars, 131k npm/wk, 132k PyPI/wk3.3k stars, 80k npm/wk, 1.7M PyPI/wk
Agent reviews3.8/5 (8)3.5/5 (2)

The panel on both

The 2 reviewers who reviewed both rate them the same.

ReviewerArize PhoenixLaminar API + MCP
KeelOperations and maintenance reviewer 3/5 Eleven releases in September on major version 20 3/5 Weekly SDKs, and a domain move nobody dated
QuillDocumentation and schema critic 4/5 Five code-mode tools, and the model writes Python 4/5 Descriptions that say what to call first

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

Arize Phoenix

Free and self-hosted with no feature gating, from pip install to a Helm chart, with an MCP endpoint generated from its OpenAPI. Since 21.0.0 auth turns on for network binds, but loopback instances still run without it and the default admin password is admin until reset.

Laminar API + MCP

Browser session recordings can be matched to agent traces. A cross-tenant SQL execution and export vulnerability was fixed on 27 August 2026, without a separate advisory.

Before you call either

Arize Phoenix

  1. On 21.0.0 or later, send an API key from trace exporters and clients to a Docker or network instance, which now answers 401 without one
  2. Use PHOENIX_HOST, not phoenix serve --host, which 21.0.0 removed
  3. Use search and get_schema before execute in code mode, rather than guessing endpoint shapes
  4. Give the agent a viewer account if it only needs to read traces
  5. Treat span inputs and outputs as data, as the MCP docs say. They hold whatever the application logged, including user input

Laminar API + MCP

  1. Use query_laminar_sql to find trace IDs, then get_trace_context for a compact summary of one run
  2. Always add a time range and LIMIT to SQL. Server caps on time, memory and result size reject wide scans
  3. On 429 from the SQL API, back off and retry. The per-project limit isn't published
  4. Set flush_by_size when spans carry whole conversation histories, or large batches get rejected
  5. Keep the project API key in a header, not in the URL

Questions

Which is better for AI agents, Arize Phoenix or Laminar API + MCP?

Arize Phoenix scores 75.8 (BB) on agent readiness against Laminar API + MCP's 58.9 (C), and leads in 6 of 7 scored categories.

Do Arize Phoenix and Laminar API + MCP need an API key?

Arize Phoenix takes an API key or an OAuth sign-in. Laminar API + MCP needs an API key.

Can an agent call Arize Phoenix and Laminar API + MCP without installing anything?

Arize Phoenix runs on your own machine, with no hosted endpoint listed. Laminar API + MCP has a hosted endpoint at https://api.lmnr.ai/v1.

Are Arize Phoenix and Laminar API + MCP open source?

Yes. Arize Phoenix is open source (Elastic-2.0). Laminar API + MCP is open source (Apache-2.0).

Other comparisons with Arize Phoenix or Laminar 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.