← All companies

Agents / Company profile

Helicone

What Helicone does

Helicone is an open-source observability and routing platform for LLM applications that aims to make production LLM behavior easier to debug, cost-control, and keep reliable across provider outages. The platform combines (1) an AI Gateway/proxy layer for routing requests across model providers and (2) a web dashboard and supporting services for end-to-end request tracing, cost/latency tracking, and workflow/session debugging for multi-step LLM and agentic flows. Helicone is designed to support developers and data scientists building production chatbots, document processing systems, and agentic pipelines, with integration options including an AI Gateway and SDK/proxy-style usage.

As of March 3, 2026, Helicone was acquired by Mintlify. Mintlify’s announcement states it is acquiring Helicone (described as an open-source LLM observability platform and AI gateway) and that Helicone’s team, including founders Justin Torre and Cole Gottdank, are joining Mintlify to help build “next generation of knowledge infrastructure.” After the acquisition, Helicone is stated to remain live in “maintenance mode” (security updates, bug fixes, and new models).

News

Company record

Mar 03, 2026 · Official · MintlifyMintlify acquires Helicone

Mintlify announced it is acquiring Helicone, described as an open-source LLM observability platform and AI gateway, and said Helicone’s team (including founders Justin Torre and Cole Gottdank) will join Mintlify in San Francisco while Helicone continues operating in maintenance mode.

Mar 03, 2026 · Official · HeliconeHelicone joins Mintlify (maintenance mode)

Helicone posted an announcement that it has been acquired by Mintlify and that Helicone’s services will remain live in maintenance mode (security updates, new models, bug and performance fixes). The post also includes usage metrics and a summary of why the founders chose the move.

Jul 22, 2025 · Official · Helicone ChangelogPrompt Management V2

Helicone’s changelog announced Prompt Management V2, describing prompt version control and composability, including deployment through the AI Gateway without rebuilding workflows.

Jun 19, 2025 · Official · Helicone ChangelogHelicone AI Gateway - Now Available

Helicone’s changelog announces the launch of Helicone AI Gateway as an open-source solution for routing, caching, and managing LLM traffic at scale, highlighting load balancing, semantic caching, failover, rate limiting, and observability integration.

May 06, 2025 · Official · Helicone ChangelogSmarter Sessions design

Helicone’s changelog announces a redesigned Sessions UI and associated session-level metrics for easier debugging of multi-step LLM workflows.

May 05, 2025 · Official · Helicone ChangelogHelicone Self-Hosting re-introduced

Helicone’s changelog states that self-hosting was re-introduced, enabling organizations to deploy Helicone in their own infrastructure for data sovereignty and compliance needs.

Apr 22, 2025 · Official · Helicone ChangelogHelicone Open WebUI integration

Helicone’s changelog announces monitoring and observability for local LLM interactions via Open WebUI, including guidance for integrating with local deployments.

Apr 07, 2025 · Official · Helicone ChangelogHelicone supports OpenAI Realtime API logging

Helicone’s changelog states it supports logging for OpenAI’s Realtime API, positioned for low-latency, multi-modal conversational experiences.

Show 2 earlier updates
Source map · 3 recurring channels · 15 references

Still resolving: Newsroom · Official X

What it builds

Helicone AI Gateway

Helicone AI Gateway is a high-performance proxy/gateway component positioned between applications and LLM providers to provide routing, caching, failover, rate limiting, and observability integration for LLM traffic.

source ↗
Helicone observability platform (dashboard/tracing)

Helicone’s platform provides LLM-specific observability, combining full request/response tracing and debugging (including session trees for multi-step workflows) with cost and reliability visibility to help reproduce and diagnose production issues.

source ↗
Helicone prompt management

Helicone offers prompt management capabilities (including prompt versioning and deployment via the AI Gateway) as part of its broader development lifecycle for LLM applications.

source ↗
Helicone self-hosting (open-source deployment)

Helicone provides an open-source, self-hosted deployment option intended for teams that want to keep LLM observability data behind their firewall/data sovereignty requirements.

source ↗

Milestones & partnerships