Master Helicone with our step-by-step tutorial, detailed feature walkthrough, and expert tips.
Create a free Helicone account at helicone.ai and generate your Helicone API key. Replace your LLM provider's base URL with Helicone's proxy URL (e.g., oai.helicone.ai for OpenAI) and add your Helicone
Auth header. Send your first LLM request through the proxy and verify it appears in the Helicone dashboard with cost and latency metrics. Add custom properties via Helicone
Property headers to segment requests by user, feature, or environment. Enable gateway features like caching, rate limiting, or retries by adding the corresponding Helicone headers to your requests.
💡 Quick Start: Follow these 3 steps in order to get up and running with Helicone quickly.
Explore the key features that make Helicone powerful for ai observability workflows.
All LLM requests are captured by routing through Helicone's gateway with zero code changes. Supports OpenAI, Anthropic, Azure OpenAI, Google, Cohere, and Mistral. Logs include full request/response bodies, latency, token counts, and computed costs.
Adding complete LLM request logging to an existing production application in under 5 minutes by changing only the API base URL — no SDK installation or code modification needed
Dashboard showing real-time spend with breakdowns by model, user, custom property, and time period. Configurable budget alerts notify when spend exceeds thresholds per day, week, or month.
Discovering that your GPT-4 usage spiked 3x this week because a new feature accidentally calls it instead of GPT-4o-mini, before the monthly bill arrives
Identical requests return cached responses from Helicone's cache layer, controlled via cache headers with configurable TTL and bucket-based caching. Cache-hit rates are tracked in the dashboard.
Reducing API costs by 40% on a FAQ chatbot where many users ask similar questions that generate near-identical API calls
Attach arbitrary key-value metadata to requests via HTTP headers (Helicone-Property-*). Properties flow through to analytics for segmentation by user, feature, environment, or any custom dimension.
Segmenting LLM costs by product feature to determine which features are most expensive to operate and which need prompt optimization
Configurable rate limits per user or API key enforced at the gateway. Automatic retry with exponential backoff for failed requests, preventing application-level retry storms.
Preventing a single power user from consuming your entire OpenAI rate limit while ensuring failed requests are retried gracefully without application code changes
Track prompt variations and model experiments with statistical significance analysis, comparing latency, cost, and quality metrics across different configurations.
Testing whether GPT-4o-mini with a longer prompt produces comparable quality to GPT-4o with a shorter prompt at 1/10th the cost, with statistical confidence
Typically 20-50ms per request based on Helicone's published benchmarks. For most applications this is negligible since LLM calls themselves take 500ms-30s — meaning the overhead represents less than 5% of total request time. For latency-critical applications making many sequential calls in agent loops, the overhead can compound and become noticeable. Helicone offers an async logging mode that bypasses the proxy entirely for teams where every millisecond counts — you send requests directly to the LLM provider and POST the request/response data to Helicone's logging endpoint afterward, eliminating any proxy overhead while still capturing full observability data.
Helicone has added session tracking that groups related requests together using a Helicone-Session-Id header, but it's primarily designed around individual request observability. You can attach session IDs and parent-child relationships via Helicone-Parent-Id headers to build hierarchical trace trees, but the visualization is less detailed than dedicated tracing platforms. For deep multi-step agent tracing with custom spans, complex tool call hierarchies, and retrieval pipeline visualization, dedicated tracing tools like Langfuse or LangSmith provide richer instrumentation through their SDK-based approaches. Helicone's strength is capturing every LLM call with minimal setup; for full agent workflow tracing, consider pairing Helicone's gateway-level logging with a dedicated tracing SDK.
Helicone focuses on operational observability (cost tracking, caching, rate limiting) with dead-simple proxy integration that takes under 5 minutes. Langfuse provides deeper tracing, evaluation, and prompt management with SDK-based integration that takes longer to set up but captures richer agent context. Helicone is the better choice when cost visibility and operational controls are the priority; Langfuse wins when you need detailed workflow tracing and evaluation pipelines for complex agent applications. The integration models differ fundamentally — Helicone's proxy approach requires no code changes beyond a URL swap, while Langfuse's decorator and callback-based SDK captures arbitrary application steps beyond just LLM calls. Many teams use both together: Helicone at the gateway for cost controls and caching, and Langfuse via SDK for deep tracing and prompt management.
Yes, Helicone is fully open-source under MIT license and can be self-hosted via Docker. The self-hosted version requires running the proxy gateway, a Supabase backend for storage and authentication, and ClickHouse for analytics, plus optional Redis for caching. It's more operationally complex than the cloud version but gives you full data control — important for healthcare, finance, and EU-based teams with data residency requirements. Helicone publishes a docker-compose setup in their GitHub repository (github.com/Helicone/helicone) with deployment documentation. The self-hosted version includes all core features: request logging, cost analytics, caching, rate limiting, and the full dashboard experience. Enterprise customers can also get dedicated support for on-premise deployments.
Helicone supports 20+ providers including OpenAI, Anthropic, Azure OpenAI, Google (Vertex AI and Gemini), AWS Bedrock, Cohere, Mistral, Groq, Together AI, Fireworks AI, OpenRouter, Perplexity, DeepInfra, Replicate, and custom model endpoints. OpenAI and Anthropic have the most seamless one-line integration via dedicated proxy URLs (oai.helicone.ai and anthropic.helicone.ai). Other providers use the universal Helicone-Target-URL header pattern, which works with any HTTP-based LLM API. Cost calculations are pre-configured for major providers and models, with automatic token counting and per-model pricing. Since the proxy simply forwards HTTP requests, adding support for new providers is straightforward — any endpoint accessible via HTTP can be routed through Helicone's gateway.
Now that you know how to use Helicone, it's time to put this knowledge into practice.
Sign up and follow the tutorial steps
Check pros, cons, and user feedback
See how it stacks against alternatives
Follow our tutorial and master this powerful ai observability tool in minutes.
Tutorial updated March 2026