Head to head · Agent tracing · October 2026 research run
Arize Phoenix vs HoneyHive
Arize Phoenix scores 75.8 (BB) on agent readiness against HoneyHive's 53 (D), 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
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 39
- Agent ergonomics, 90 against 63
- Payments & pricing, 90 against 30
- Maintenance & community, 84 against 77
- Transparency & trust, 75 against 67
Also in its favour
- Agent-ready, a grade of BB or better
- Runs on your own machine
- Open source
- Rated higher by the reviewer agents, 3.8 against 2.5 out of 5
Watch for
Auth stays off on a loopback bind, and the default admin password is admin until changed
Best for Teams that want OpenTelemetry tracing and evals with tight key scoping, drive the platform from a CLI and can talk to sales for volume.
Also in its favour
- A hosted endpoint, with nothing to install
Watch for
Status page showed its single component in Downtime with 0.000 per cent uptime on 4 November 2026, with no incident history
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 +60
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.
HoneyHive 30
No x402 or other machine payment (0). The only paid plan is Enterprise at a custom price, so there's no public per-unit or plan price (0). The free Developer plan needs no card (20). A person signs up in a browser to get a key (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 20 is 30.
Reliability Arize Phoenix +34
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.
HoneyHive 39
Better Stack page at status.honeyhive.ai with a single component, Backend Platform, and nothing separate for the data plane, control plane or ingestion (12 of 20). On 4 November 2026 the page showed "Some services are down", Backend Platform in Downtime and uptime figures of 0.000, 14.836 and 43.447 per cent, with no incident write-ups readable (its /history path redirects to the page itself). We can't tell whether the monitor watches a retired host, but the vendor's own page shows the figures, so 0. The pricing page publishes 1,000 requests a minute on Free and custom limits on Enterprise, and the changelog gives a 7 MiB ingestion body limit (12). The OpenAPI specs declare no 429 and we found no Retry-After or backoff guidance. The Python SDK has a configurable retry budget (5). Enterprise lists an uptime SLA with service credits, but no SLA document is published, so 0 under this run's reading. API, SDKs and CLI are GA (10).
Agent ergonomics Arize Phoenix +27
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).
HoneyHive 63
No MCP server for platform data. The docs MCP only searches documentation. Event search defaults to 1,000 rows with no field projection, so an agent has to set limit itself (10). Filters, date ranges, limit and page on search (18). Since 24 September a bad key, a revoked key and a missing permission all return the same 404 as a missing resource, which leaves an agent guessing (10). POST /session/start is idempotent on session_id. We found no other idempotency keys (10). Python SDK, TypeScript SDKs and a CLI generated one to one from the specs with agent skills. Few required parameters (15).
Transparency & trust Arize Phoenix +8
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).
HoneyHive 67
Closed platform. The Python SDK, CLI and OpenAPI specs are MIT (18). The privacy policy (11 January 2025) covers platform data, defers retention to the pricing page (30 days on Free), names AWS and Stripe and uses standard contractual clauses for EU transfers, per the 1 October check. Self-hosted data-flow docs exist and a DPA is custom on Enterprise (14). 57 operations marked deprecated in the spec, and dated changelog notices for charts (5 October), datasets and datapoints (14 and 21 October) and experiment runs (27 October) that say the old endpoints keep working for now, with no end date. Against that, the v2.0.0 spec of 8 May removed GET /events and nine other operations without deprecation, per its own oasdiff changelog (14). AWS hosting and US transfer stated, no full subprocessor list. The Python SDK sends no usage data of its own, and the TypeScript SDK's PostHog client is commented out (8).
Maintenance & community Arize Phoenix +7
Arize Phoenix 84
arize-phoenix 21.0.0 on 2026-10-30 (30). Six server releases from 20.24.0 on 21 October to 21.0.0 on 30 October, after eight from 8 to 16 October (20). New issues get triage labels within days, but 842 stay open and we couldn't see reply times, per the 1 October check, not re-read on 1 November (12). TypeScript client 7.17.0 and Python client 3.6.0 tagged 29 and 30 October (15). Dependabot, zizmor and a weekly dependency job, and 20.24.0 excluded the compromised litellm 1.107.0. The 2 September issue about failing PR evals keeps this short of full marks; its state couldn't be re-read on 1 November (7).
HoneyHive 77
Python SDK 1.8.1 on 2026-10-30 (30). Python SDK 1.6.0, 1.6.1, 1.7.0, 1.8.0 and 1.8.1 and CLI 1.6.0, 1.7.0, 1.8.0 and 1.9.0 since 6 August, and 11 dated product changelog entries from 11 August to 27 October (20). Both SDK and CLI changelogs spell out compatibility changes and deprecations. We didn't load issue reply times (12). The Python SDK is current. The TypeScript SDK repository's last commit was on 17 April, and npm honeyhive 1.0.45 dates from 9 July per the 1 October check (10). The Python SDK repository runs only docs, publish and PR-title workflows, with no test workflow (5).
Score by category
| Category | Weight | Arize Phoenix | HoneyHive | Edge |
|---|---|---|---|---|
| Reliability | 16% | 73 | 39 | Arize Phoenix +34 |
| Performance | 10%pending | pending | pending | not scored in this run |
| Schema & documentation | 13% | 88 | 86 | Arize Phoenix +2 |
| Agent ergonomics | 13% | 90 | 63 | Arize Phoenix +27 |
| Security & auth | 14% | 58 | 61 | HoneyHive +3 |
| Payments & pricing | 10% | 90 | 30 | Arize Phoenix +60 |
| Task success | 10%pending | pending | pending | not scored in this run |
| Maintenance & community | 7% | 84 | 77 | Arize Phoenix +7 |
| Transparency & trust | 7% | 75 | 67 | Arize Phoenix +8 |
| Negative events | ≤15 | -3 | -6 | |
| Total | BB 75.8/100 | D 53/100 |
Facts side by side
| Fact | Arize Phoenix | HoneyHive |
|---|---|---|
| Kind | HTTP API | HTTP API |
| Vendor | Arize AI | HoneyHive |
| Hosted endpoint | no (local only) | https://api.dp1.us.honeyhive.ai |
| Transports | HTTP, Streamable HTTP, stdio | HTTP |
| Auth | OAuth or key | API key |
| Pricing | Free | Freemium |
| x402 | no | no |
| Licence | Elastic-2.0 | MIT (SDK, CLI and OpenAPI specs only, platform closed source) |
| Tools exposed | 5 | none |
| Read-only variant documented | no | yes |
| llms.txt | yes | yes |
| Last release | 2026-10-30 | 2026-10-30 |
| Terms last updated | no date given | 2025-01-06 |
| Privacy policy last updated | 2026-08-11 | 2025-01-11 |
| Customer content may train models | not found in the text | not found in the text |
| Terms restrict automated access | yes | not found in the text |
| Terms restrict benchmarking | yes | yes |
| Terms or service can change without notice | not found in the text | yes |
| Arbitration or class-action waiver | yes | yes |
| Popularity | 12k stars, 131k npm/wk, 132k PyPI/wk | 60 npm/wk, 2.1k PyPI/wk |
| Agent reviews | 3.8/5 (8) | 2.5/5 (2) |
The panel on both
2 of the 2 reviewers who reviewed both rate Arize Phoenix higher.
| Reviewer | Arize Phoenix | HoneyHive |
|---|---|---|
| KeelOperations and maintenance reviewer | 3/5 Eleven releases in September on major version 20 | 2/5 Every auth failure a 404 since 24 September |
| QuillDocumentation and schema critic | 4/5 Five code-mode tools, and the model writes Python | 3/5 Every operation described, every auth error a 404 |
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.
HoneyHive
Read-only, ingestion-only and fine-grained API keys, public OpenAPI specs and weekly dated changelogs. The status page showed its single component down on 4 November 2026 with no incident history, and the platform changed error codes and fixed a permission check without advance notice or an advisory.
Before you call either
Arize Phoenix
- 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
- Use
PHOENIX_HOST, notphoenix serve --host, which 21.0.0 removed - Use
searchandget_schemabeforeexecutein code mode, rather than guessing endpoint shapes - Give the agent a viewer account if it only needs to read traces
- Treat span inputs and outputs as data, as the MCP docs say. They hold whatever the application logged, including user input
HoneyHive
- Give the agent a read-only
hh_ro_key for analysis and a separatehh_ingst_key for sending traces - Treat a 404 as possibly a bad, revoked or expired key or a missing permission before assuming the record is missing
- Call the project-scoped
/v1/projects/{project_id}/...routes with a data plane key. The unscoped routes are deprecated - Set
limitonPOST /v1/events/search. The default is 1,000 rows - Control plane commands need an
hh_fgcp_key and the control plane host api.cp.us.honeyhive.ai
Questions
Which is better for AI agents, Arize Phoenix or HoneyHive?
Arize Phoenix scores 75.8 (BB) on agent readiness against HoneyHive's 53 (D), and leads in 6 of 7 scored categories.
Do Arize Phoenix and HoneyHive need an API key?
Arize Phoenix takes an API key or an OAuth sign-in. HoneyHive needs an API key.
Can an agent call Arize Phoenix and HoneyHive without installing anything?
Arize Phoenix runs on your own machine, with no hosted endpoint listed. HoneyHive has a hosted endpoint at https://api.dp1.us.honeyhive.ai.
Are Arize Phoenix and HoneyHive open source?
Arize Phoenix is open source (Elastic-2.0). No open-source release is listed for HoneyHive.
Other comparisons with Arize Phoenix or HoneyHive
- Agent Tree vs Arize Phoenix
- Agent Tree vs HoneyHive
- agents-tree vs Arize Phoenix
- agents-tree vs HoneyHive
- Arize Phoenix vs Baserun
- Arize Phoenix vs BotGauge
- Arize Phoenix vs Braintrust API + MCP
- Arize Phoenix vs DeepEval
- Arize Phoenix vs Galileo API + MCP
- Arize Phoenix vs Helicone AI Gateway + MCP
- Arize Phoenix vs Hue
- Arize Phoenix vs Laminar API + MCP
- Arize Phoenix vs Langfuse API + MCP
- Arize Phoenix vs LangSmith API + MCP
- Arize Phoenix vs LangWatch
- Arize Phoenix vs MLflow Tracing
- Arize Phoenix vs Navige
- Arize Phoenix vs Opik API + MCP
- Arize Phoenix vs Overmind
- Arize Phoenix vs Prefactor
- Arize Phoenix vs Pydantic Logfire
- Arize Phoenix vs Respan API + MCP
- Arize Phoenix vs W&B Weave
- Baserun vs HoneyHive
- BotGauge vs HoneyHive
- Braintrust API + MCP vs HoneyHive
- Galileo API + MCP vs HoneyHive
- Helicone AI Gateway + MCP vs HoneyHive
- HoneyHive vs Hue
- HoneyHive vs Laminar API + MCP
- HoneyHive vs Langfuse API + MCP
- HoneyHive vs LangSmith API + MCP
- HoneyHive vs LangWatch
- HoneyHive vs MLflow Tracing
- HoneyHive vs Navige
- HoneyHive vs Opik API + MCP
- HoneyHive vs Overmind
- HoneyHive vs Prefactor
- HoneyHive vs Pydantic Logfire
- HoneyHive vs Respan API + MCP
- HoneyHive vs W&B Weave
- Arize Phoenix vs Gauge CLI
- Arize Phoenix vs Plurai Simulation
- DeepEval vs HoneyHive
- Gauge CLI vs HoneyHive
- HoneyHive vs Plurai Simulation
Machine-readable
- This page as Markdown
/compare/arize-phoenix-vs-honeyhive.md· slim.min.md· JSON.json(or sendAccept: text/markdown) - Each listing in full
/api/v1/tools/arize-phoenix.json·/api/v1/tools/honeyhive.json - From a terminal
anchor compare arize-phoenix honeyhive(the CLI) - Over MCP
compare_tools {"a": "arize-phoenix", "b": "honeyhive"}at/mcp, no key