Helicone

AI gateway and LLM observability platform to route, debug, and analyze AI apps.

Pricing

Hobby free (10,000 requests/month); Pro $79/month with 7-day trial; Team $799/month; Enterprise custom. Joined Mintlify (announced on site).

Overview

Helicone is an AI gateway and observability layer for LLM and agent applications. It logs requests, supports sessions and user analytics, prompts testing, caching, rate limits, and automatic fallbacks across providers. Teams use HQL for querying usage data and playgrounds for prompt iteration. Pricing scales from a free Hobby tier through Pro, Team, and Enterprise (including on-prem).

Key features

Request logging, sessions, and user analytics
AI gateway with caching, rate limits, and fallbacks
Prompt playground, scores, and datasets
HQL query language for usage analysis
Alerts and reports on paid plans
SOC 2 / HIPAA options on Team+
On-prem and SSO on Enterprise

Use cases

Show all 7Show fewer

Alternatives

AgentOpsDeveloper observability platform to trace, debug, and cost-track ai agents.AI toolFreemium
ArizeObservability and evaluation platform for tracing, testing, and improving ai agents.AI toolFreemium
Azure AI FoundryMicrosoft Azure platform to build, deploy, and govern AI apps and agents with models and tools.AI toolPaid
FireworksInference and training platform for open models, fine-tunes, and production LLM serving.AI toolFreemium
Hugging FaceOpen ML platform for models, datasets, Spaces apps, and inference tooling.AI toolFreemium
IBM GraniteApache 2.0 open IBM foundation models for language, speech, vision, embeddings, and guardrails.AI toolFree

Last updated 2026-09-16