Both changed hands in 2026, with different results. Helicone has been in maintenance mode since March 2026 and is not taking new cloud sign-ups, and on 19 October 2026 it said it is no longer part of Mintlify, with no new owner named. Arize continues to ship Phoenix and AX under Dynatrace, so it is the one a new team can actually adopt.[3][6]
Arize is an evaluation and observability vendor with a free self-hosted tool, Phoenix, and a hosted platform, AX; Helicone is an LLM request logger and AI gateway that has been in maintenance mode since March 2026, with new cloud sign-ups disabled. Dynatrace completed its acquisition of Arize on 1 October 2026 and says Phoenix and AX continue. For a new deployment Arize is usually the practical choice; existing Helicone users mainly need a migration plan.
| Decision factor | Arize | Helicone |
|---|---|---|
| Ownership | Acquired by Dynatrace, completed 1 October 2026; Phoenix and AX continue | Acquired by Mintlify in March 2026; on 19 October 2026 said it is no longer part of Mintlify, no new owner named |
| Status | Active: Phoenix and AX releases through October 2026 | Maintenance mode since March 2026: patches, bug fixes and model support only |
| New cloud accounts | Yes, AX Free and paid plans | No; sign-ups disabled since March 2026 |
| Centre of gravity | Evaluation and tracing: datasets, experiments, judges, online evals in AX | Request logging, cost tracking and an AI gateway |
| Licence | Phoenix: Elastic License 2.0 (source-available); AX: commercial | Apache-2.0; separate Rust gateway is GPL-3.0, last pushed November 2025 |
| Gateway | Phoenix proxy (August 2026) not positioned for production; AX integrates third-party gateways | AI Gateway for 100+ providers with failover, caching and rate limits |
| Free tier | AX Free: 25k spans a month, 15-day retention; Phoenix free to self-host | Hobby: 10K requests, 7-day retention, if you already have an account |
| Capability | Arize | Helicone | Prefactor |
|---|---|---|---|
| Observe | |||
| LLM call tracing | ✓ | ✓ | ✓ |
| Multi-step agent tracing (tool calls, nested spans) | ✓ | Partial | ✓ |
| Cost tracking | ✓ | ✓ | ✓ |
| OpenTelemetry support | ✓ | Partial | ✓ |
| Evaluate | |||
| Offline evals (datasets, experiments) | ✓ | Partial | ✓ |
| Online evals on production traffic | ✓ | Partial | ✓ |
| LLM-as-a-judge scoring | ✓ | Partial | ✓ |
| Code-based evaluators | ✓ | — | ✓ |
| Human review and annotation | ✓ | Partial | ✓ |
| Prompt versioning and playground | ✓ | ✓ | — |
| Act on findings | |||
| Runtime controls on live agent actions | Partial | Partial | ✓ |
| Gateway or proxy in the request path | Partial | ✓ | — |
| Licence and deployment | |||
| Open source | Partial | ✓ | Partial |
| Self-hosting | ✓ | ✓ | Partial |
Mintlify announced its acquisition of Helicone on 3 March 2026 and moved the product into a maintenance-only phase: patches, bug fixes and new model support, plus help for customers moving to other platforms. The cloud sign-up page has shown 'Sign ups are disabled' since the same day, and the changelog shows no feature entries after November 2025. On 19 October 2026 Helicone's site began saying it is no longer part of Mintlify, with details to come and no new owner named.
Arize's ownership changed too: Dynatrace completed its acquisition on 1 October 2026. Its announcement says Arize will keep supporting both Phoenix and AX and will integrate Arize capabilities into Dynatrace over time. Phoenix and AX both shipped releases in October 2026. For a team starting now, Helicone is not an option on its cloud, and Arize is.
Helicone also had a four-day outage of its dashboard and log processing in May 2026, when AWS locked its account; the proxy stayed up and queued logs were processed afterwards.
The two were never the same kind of product. Helicone logs every model request through its proxy or async SDKs, groups requests into sessions using headers, tracks cost per request and session, and routes traffic through an OpenAI-compatible gateway with failover and caching. Its own documentation says it is not an evaluation framework. Python evaluators were turned off in October 2026, and the Experiments feature was deprecated in 2025.
Arize is built around evaluation. Phoenix, free to self-host, covers tracing, datasets, experiments, LLM and code evaluators and prompt versioning. AX, the hosted platform, adds continuous online evals on production traces, monitors, alerting and an agent trajectory graph. Phoenix's documentation points to AX for continuous production evaluation.
Cost tracking is where Helicone was strongest: exact per-request cost through its gateway's model registry, pricing for more than 300 models, and cost alerts with graduated thresholds. Arize covers cost too. Phoenix calculates it from token counts and a built-in pricing table at span, trace, session and project level, and AX added a Cost tab on every project in September 2026.
Helicone's async logging is built on OpenLLMetry instrumentation through its own SDKs. Arize is built on OpenTelemetry and maintains the OpenInference conventions, so the move means re-instrumenting with OpenInference or another OpenTelemetry route that Arize documents, then rebuilding dashboards and cost views in Phoenix or AX.
The gap to plan for is the gateway. If your application sends model traffic through Helicone's AI Gateway for routing, failover or caching, Arize does not replace that: Phoenix's chat completions proxy does not support tool calling and is not positioned for production, and AX integrates with third-party gateways such as Portkey and Gravitee. Self-hosted Helicone is Apache-2.0 and keeps running, but it receives only maintenance changes, and its Docker image no longer proxies LLM traffic without the separate gateway.
Treat prompts and feedback as separate moves. Prompts versioned in Helicone need recreating in Phoenix's prompt versioning or in AX's Prompt Hub, which supports diffing, rollback and environment tags. Helicone's positive and negative feedback on requests maps onto annotations in Phoenix, or onto AX annotation queues.
As of 20 October 2026, Arize AX Free is $0 for 25k spans a month with 15-day retention and unlimited users, and AX Pro is $50 a month for 50k spans with 30-day retention. AX Enterprise is custom and is the only self-hosted AX option. Phoenix is free to self-host.
Helicone lists Hobby free with 10K requests and 7-day retention, Pro at $79 a month and Team at $799 a month. Because cloud sign-ups are disabled, those prices apply only to existing accounts. Check Arize's pricing and Helicone's pricing for current terms.
AX includes unlimited users and unlimited evals and experiments on every plan.
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 for risky actions, and a kill switch for a run, an agent or a team. Arize AX documents guards that run in your own code around model calls, and Helicone's gateway can block prompt injection for OpenAI models only; neither holds an agent's tool action for approval. Prefactor has no prompt playground or prompt registry, no gateway in the request path, and its platform is hosted, so if free self-hosting or a gateway replacement is the requirement, look elsewhere. 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.