Both trace agents on OpenTelemetry, run experiments and score production traffic. Arize, now owned by Dynatrace, gives you a free self-hosted tool and published hosted tiers; HoneyHive is an independent, seed-funded hosted platform whose paid plans are all sales-led. How you want to buy and where you want it to run usually settles it.[3][4]
Arize offers Phoenix, a free source-available tool you run yourself, and Arize AX, a hosted platform with self-serve paid tiers; HoneyHive is a hosted platform with a free Developer tier and quote-based Enterprise pricing above it. Dynatrace completed its acquisition of Arize on 1 October 2026, and Phoenix and AX continue. HoneyHive is independent, and both are built on OpenTelemetry.
| Decision factor | Arize | HoneyHive |
|---|---|---|
| Ownership | Acquired by Dynatrace, completed 1 October 2026; Phoenix and AX continue | Independent; $7.4M total funding last announced April 2025 |
| Products | Phoenix (self-hosted) and Arize AX (hosted or Enterprise self-hosted) | One hosted platform, with Enterprise self-hosting |
| Free self-hosted option | Phoenix, under the Elastic License 2.0 | None; self-hosting is Enterprise only, documented on AWS |
| Online evals | AX only; Phoenix docs point to AX | Yes, sampled LLM, code and human evaluators |
| Run-time controls | Partial: AX guards run in your code via Guardrails AI | None; Policies set evaluation standards, not enforcement |
| Notable integrations | OpenInference instrumentors; agent trajectory graph for LangGraph, CrewAI, AG2 | Salesforce Agentforce, Databricks Agent Bricks, Copilot Studio, coding agents |
| Free hosted tier | AX Free: 25k spans a month, 15-day retention, unlimited users | Developer: 10,000 events a month, 30-day retention, up to 5 users |
| First paid tier | AX Pro: $50 a month, 50k spans | No self-serve paid plan; Enterprise is quote-based |
| Capability | Arize | HoneyHive | Prefactor |
|---|---|---|---|
| Observe | |||
| LLM call tracing | ✓ | ✓ | ✓ |
| Multi-step agent tracing (tool calls, nested spans) | ✓ | ✓ | ✓ |
| Cost tracking | ✓ | ✓ | ✓ |
| OpenTelemetry support | ✓ | ✓ | ✓ |
| Evaluate | |||
| Offline evals (datasets, experiments) | ✓ | ✓ | ✓ |
| Online evals on production traffic | ✓ | ✓ | ✓ |
| LLM-as-a-judge scoring | ✓ | ✓ | ✓ |
| Code-based evaluators | ✓ | ✓ | ✓ |
| Human review and annotation | ✓ | ✓ | ✓ |
| Prompt versioning and playground | ✓ | ✓ | — |
| Act on findings | |||
| Runtime controls on live agent actions | Partial | — | ✓ |
| Gateway or proxy in the request path | Partial | — | — |
| Licence and deployment | |||
| Open source | Partial | Partial | Partial |
| Self-hosting | ✓ | ✓ | Partial |
Dynatrace completed its acquisition of Arize on 1 October 2026. Its announcement says Arize will keep supporting both Phoenix and AX and that Arize capabilities will be integrated into Dynatrace over time. For Dynatrace customers that points towards agent evaluation inside a tool they already run. For everyone else, expect packaging and pricing to shift over the next few quarters.
HoneyHive has not changed hands. Its last announced funding was $7.4M in total, including a seed round led by Insight Partners in April 2025. It has kept a steady release pace through 2026: workspaces in January, annotation queue automations in April, coding-agent tracing in June, Agentforce and Databricks tracing in August and a Conversation view in September. In June 2026 it was named the observability partner in Microsoft's ASSERT and Agent Control Specification announcement.
Arize has two free routes. Phoenix is free to self-host with tracing, datasets, experiments, LLM and code evaluators and prompt versioning, but continuous evaluation of production traffic, monitors and alerting are AX features. Phoenix uses the Elastic License 2.0: you can modify and self-host it, but not offer it to others as a hosted service. AX Free includes 25k spans a month, 1 GB of ingestion and 15-day retention with unlimited users, and AX Pro is $50 a month for 50k spans, 10 GB and 30-day retention, as of 20 October 2026.
HoneyHive's Developer tier is free with no card, 10,000 events a month, 30-day retention, up to five users and one workspace. There is no published self-serve paid plan: beyond the free tier you talk to sales about Enterprise, which adds custom limits, unlimited users and workspaces, SSO and dedicated, hybrid or self-hosted deployment. Online evaluations are available without moving to a separate product.
If you want to pay by card and scale gradually, Arize has the clearer route. If you expect to sign a negotiated contract anyway, the missing self-serve tier matters less. Compare Arize's pricing and HoneyHive's pricing.
Both are built on OpenTelemetry. Arize maintains OpenInference, a set of conventions and instrumentors for LLM and agent spans, and the same instrumentation works with Phoenix and AX. HoneyHive accepts OTLP and reads the OTel GenAI, OpenLLMetry and OpenInference conventions, so instrumentation written for one can usually be pointed at the other.
The integration lists differ in emphasis. HoneyHive traces enterprise agent platforms such as Salesforce Agentforce, Databricks Agent Bricks and Microsoft Copilot Studio, and coding agents including Claude Code, Codex CLI and Devin. Its full tracing SDK is Python; the TypeScript package is a REST client, so TypeScript teams send OTLP instead. Arize ships Python and TypeScript SDKs, a Go client for Phoenix and OpenInference instrumentors in Python, JavaScript and Java. Its AX Agent Trajectory view draws which agents call which, set automatically for LangGraph, CrewAI and AG2.
Both track cost. Arize calculates it from token counts at span, trace and session level and added a Cost tab to every AX project in September 2026; HoneyHive offers cost, token and latency dashboards and alerts, with usage reports by workspace and project since July 2026. On compliance, AX lists SOC 2 Type II, ISO 27001, HIPAA and PCI DSS 4.0, while HoneyHive lists SOC 2 Type II, GDPR with EU data residency, and HIPAA with a BAA on Enterprise.
Arize AX documents guards built on the Guardrails AI framework that run in your own code around model calls and can return a default response, re-ask or raise an exception. HoneyHive documents no blocking or approvals; its Policies feature sets organisation-wide evaluation and monitoring standards, and in the Microsoft partnership HoneyHive is described as the observability layer while the Agent Control Specification does the enforcing. Neither offers approval holds on agent tool calls.
We sell the layer this section describes. Read it with that in mind.
Stated plainly, since Prefactor publishes this page: Prefactor is an agent evaluation platform that scores each step of a run and acts on what it finds through runtime policies on tool calls and data access, approval holds and a kill switch scoped to a run, an agent or a team. Arize's guards cover model calls in your own code, and HoneyHive leaves enforcement to other tools. Prefactor has no prompt playground or prompt registry and its platform is hosted, so if free self-hosting matters most, Phoenix is the better fit. See the full comparison of 15 platforms.
Reviewed against public product and documentation pages on October 20, 2026. If a vendor has changed a feature, product name, or positioning since then, send a correction and we will update it. Numbered source links in the page body point to the ordered sources below.
Prefactor scores each step of every agent run and can hold or stop an action when a score says it should not happen. Start on the free tier: 5,000 spans a month, no credit card required.
Prefactor helps teams observe, evaluate, and improve their AI agents in production, across every framework and provider.