These are different kinds of product at different stages. Galileo, now Splunk Agent Observability, is an active evaluation platform with guardrails that act during a run; Helicone is a request logger and gateway in maintenance mode, with new cloud sign-ups disabled since March 2026. For a new deployment the question is usually not which to pick but where Helicone users should go next.[3][6]
Galileo is an evaluation and guardrail platform for agents, now sold as Splunk Agent Observability, while Helicone is an open-source request logger and AI gateway that has been in maintenance mode since March 2026. Helicone's cloud has not accepted new sign-ups since then, and on 19 October 2026 it said it is no longer part of Mintlify, with no new owner named. For a new deployment Galileo, or another active platform, is usually the practical choice; existing Helicone users should plan a migration.
| Decision factor | Galileo | Helicone |
|---|---|---|
| Status | Active; Cisco completed its acquisition in May 2026, sold as Splunk Agent Observability since 7 August 2026 | Maintenance mode since March 2026; on 19 October 2026 said it is no longer part of Mintlify, no new owner named |
| New customers | Yes, through Splunk | Cloud sign-ups disabled since March 2026 |
| What it is | Evaluation, observability and run-time guardrails for agents | LLM request logging, cost tracking and an AI gateway; by its own docs, not an evaluation framework |
| Agent tracing | Sessions, traces and spans; agent framework integrations and A2A tracing | Header-based session grouping, not a span model for agent steps |
| Run-time checks | Agent Control checks model and tool inputs and outputs during a run | Prompt-injection and jailbreak blocking, OpenAI models only; rate limits |
| Gateway | None | Yes: OpenAI-compatible endpoint for 100+ providers with failover and caching |
| Licence | Proprietary platform; SDKs and Agent Control are Apache-2.0 | Apache-2.0; separate Rust gateway is GPL-3.0 |
| Listed pricing | From $100 a month per 1.2 million spans a year, billed annually | Hobby free, Pro $79 a month, Team $799 a month, though new accounts cannot be opened |
| Capability | Galileo | 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 | Partial | ✓ | — |
| Act on findings | |||
| Runtime controls on live agent actions | ✓ | Partial | ✓ |
| Gateway or proxy in the request path | — | ✓ | — |
| Licence and deployment | |||
| Open source | Partial | ✓ | Partial |
| Self-hosting | ✓ | ✓ | Partial |
Mintlify announced its acquisition of Helicone on 3 March 2026 and moved the product to a maintenance-only phase covering patches, bug fixes and new model support, with help for customers to migrate to other platforms. Cloud sign-ups have shown 'Sign ups are disabled' since the same day, and the changelog has 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 follow and no new owner named.
Galileo also changed hands, but in the other direction. Cisco completed its acquisition in May 2026 and the product is sold as Splunk Agent Observability since 7 August 2026, with new features still shipping, such as annotation queues in August and metric access control in October.
So for a team choosing today, Helicone's cloud is not an option for a new account. Its Apache-2.0 code can still be self-hosted, but you would be running software whose own maintainers describe it as in maintenance.
Helicone sits in the request path. Its proxy and AI Gateway log every model request, track cost per request and per session, and route traffic across more than 100 providers with failover, caching and rate limits. Its documentation says plainly that it does not run evaluations for you; scores from external tools can be posted to requests, and its Python evaluators were turned off in October 2026.
Galileo is an evaluation platform. It runs experiments on datasets, scores production traces with LLM judges, Luna-2 small models or code metrics, and supports human annotation. Agent Control checks model and tool inputs and outputs during a run and can block prompt injection, personal data leaks and harmful content. Helicone's prompt-injection blocking works with OpenAI models only. Galileo has no gateway, so it does not replace Helicone's routing or caching.
Instrumentation differs too. Helicone's async logging is built on OpenLLMetry through its own SDKs, and it documents no generic OTLP endpoint. Galileo accepts OTLP and lists native integrations for the OpenAI Agents SDK, LangGraph, CrewAI, Google ADK, Strands and Microsoft Agent Framework, plus A2A multi-agent tracing added in May 2026. Helicone's record this year also matters for planning: a May 2026 AWS account lock took its dashboard and log processing down for about four days, while its proxy stayed up.
Existing Helicone users have two jobs, not one. The first is logging and evaluation: Galileo accepts OTLP traces and OpenInference instrumentors in Python, and has SDK integrations for the OpenAI Agents SDK, LangGraph, CrewAI and others, so instrumentation replaces Helicone's base URL or async logger. Helicone's sessions are grouped by headers, so expect to rebuild agent traces as proper spans rather than map them across.
The second is the gateway. If you rely on Helicone for provider routing, failover, caching or rate limits, Galileo does not offer those; plan for a separate gateway. Helicone's prompt versioning and stored prompts would also need a new home, since Galileo's prompts page does not document version control.
The ownership notice gives no timeline, and Mintlify's March announcement already offered help to migrate, so planning the move now costs little and avoids doing it under pressure later.
As of 20 October 2026, Helicone still lists Hobby free with 10K requests, Pro at $79 a month and Team at $799 a month, but new cloud accounts cannot be created on any tier. Splunk lists Agent Observability from $100 a month per 1.2 million spans a year, billed annually, with no free tier listed. See Splunk's pricing and Helicone's pricing; for current Helicone customers the more useful number is the cost of the platform you move to.
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. Galileo's Agent Control also blocks content during a run; Helicone's checks are limited to prompt-injection filtering on OpenAI models. Neither documents approval holds. Prefactor has no prompt playground or prompt registry, no gateway, and its platform is hosted, so Helicone users who need routing will still need a separate gateway. 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.