Introduction
Helicone is a production-ready observability, monitoring, and analytics platform for large language models (LLMs), specifically designed to track and manage AI applications in real-world deployments.
Deployed between applications and LLM services, it automatically records the complete context of every model invocation—including prompts, model responses, token usage, call latency, error messages, and actual costs.
By providing real-time dashboards, request-level tracing, user and session analytics, cost and performance breakdowns, Helicone empowers engineering teams to rapidly pinpoint issues, debug prompts, evaluate model performance, control inference costs, and continuously optimize and stabilize LLM-driven systems.
Setup Helicone AI Gateway
• Create an account
Please register an Helicone account and log in first, then continue with the setup process.
• Generate an API Key
- 1. Click “Configure” in the bottom-left corner to go to Settings
- 2. Select the API keys tab and generate a Helicone Key.
- 3. Generate a new key. You may choose whether to enable read permissions. A Helicone API key with read permissions can:
- • Read request logs
- • View statistics
- A Helicone API key with write permissions can:
- • Write request logs (most critical)
- • Function as a proxy
• Select your preferred proxy provider
- 1. Choose your preferred proxy provider from the options below and enter the provider's API Key you have on hand.
• Send Your API Requests
- 1. You can view your requests in the Requests tab.
- 2. Detailed data for your requests will be displayed in chart form on the Dashboard.
Obtain Canopy Wave API Key
Simply log in to Canopy Wave Cloud to easily obtain your Canopy Wave API Keys. Visit Canopy Wave
1. Log in to Canopy Wave Cloud
Canopy Wave Cloud supports multiple login methods for your convenience:
- Quick login with Google account
- Quick login with GitHub account
- Email registration login
2. Manage Canopy Wave API Key
- You can manage your Canopy Wave API Key under Model API → Model API Key
- Click New Model API Key to generate a new API Key
Canopy Wave API on Helicone
Canopy Wave is now officially available as an API Provider. When using Helicone, you can select Canopy Wave as your API Provider to enjoy both Helicone's precise monitoring and Canopy Wave's stable, fast model inference services!
