ai-observability
Helicone
Open-source LLM observability — 1-line integration via proxy
LLM observability via proxy or async logging. 1-line integration. Free 10k req/mo + Pro $79/mo + Team $799/mo. HQL SQL query language. SOC-2 + HIPAA on Team. MIT-licensed.
Pricing
| Tier | Price | Notes |
|---|---|---|
| Hobby (Free) | Free | 10,000 requests/month. 7-day retention. 1 seat. Basic monitoring. |
| Startup Discount | Free | <2 years, <$5M funding: 50% off first year. |
| Self-Hosted (OSS) | Free | MIT-licensed. Run Helicone yourself for free. |
| Pro | $79/mo | $79/month. 10k free + usage-based. Unlimited seats. Alerts, reports, HQL query language. 1-month retention. |
| Team | $799/mo | $799/month. 5 orgs, SOC-2 + HIPAA compliance, dedicated Slack, 3-month retention. |
| Enterprise | Custom | Custom MSA, SAML SSO, on-prem deploy, bulk discounts, forever retention. |
Limits
| Tier | Metric | Value | Notes |
|---|---|---|---|
| — | caching | Built-in prompt caching to save costs | LLM caching |
| — | free for oss | $100 annual credit for OSS projects | OSS credits |
| — | free for students | Students: free access | Student pricing |
| — | integration methods | Proxy (OpenAI-compatible base URL swap) OR async logging SDK | Integration paths |
| — | open source | Full platform open-source (MIT) | OSS license |
| — | pii redaction | Optional PII scrubbing at ingest | Privacy |
| — | providers supported | OpenAI, Anthropic, Google Gemini, AWS Bedrock, Azure OpenAI, Groq, OpenRouter, Together, Fireworks, Deepseek, Cohere, any OpenAI-compat | LLM providers |
| — | proxy latency | <10ms added latency on proxy mode | Proxy overhead |
| — | rate limiting | Per-user rate limits via proxy | Rate limiting |
Features
- Alerts — Thresholds on error rate, latency, cost, usage. Pro+.
- Async Logging — Log AFTER the LLM call via SDK — zero added latency.
- Cost Tracking — Automatic cost calculation per call by provider/model.
- Dashboard — Request tables, aggregate metrics, cost breakdowns.
- Evaluators — LLM-as-judge + custom evaluators on runs.
- Experiments — A/B test different models/prompts.
- HQL (SQL over traces) — Query your logged data with SQL. Pro+. · docs
- PII Redaction — Automatically scrub emails, credit cards, etc. from logs.
- Prompt Caching — Cache identical requests → save money.
- Prompts & Versions — Store + version + A/B test prompts.
- Proxy Mode — 1-line integration via base URL swap. Captures all requests.
- Rate Limiting — Per-user + per-key rate limit policies.
- Reports — Scheduled email reports with KPIs.
- Self-Hosting — Docker + k8s deployment.
- Sessions — Group related calls (chat sessions, agent runs).
- User Metrics — Per-user cost + usage segmentation.
Developer interfaces
| Slug | Name | Kind | Version |
|---|---|---|---|
| async-logging | Async Logging API | rest | v1 |
| cli | Helicone CLI | cli | 0.x |
| dashboard | Helicone Dashboard | other | — |
| sdk-node | helicone (npm) | sdk | 1.x |
| proxy | Helicone Proxy | rest | v1 |
| sdk-python | helicone-python | sdk | 1.x |
| rest-api | Query API (HQL) | rest | v1 |
| webhooks | Webhooks | other | — |
Compare Helicone with
ai-api
Helicone vs Anthropic API
Side-by-side breakdown.
ai-api
Helicone vs AssemblyAI
Side-by-side breakdown.
ai-api
Helicone vs Deepgram
Side-by-side breakdown.
ai-api
Helicone vs ElevenLabs
Side-by-side breakdown.
ai-api
Helicone vs Google Gemini API
Side-by-side breakdown.
ai-api
Helicone vs Groq
Side-by-side breakdown.
ai-api
Helicone vs OpenAI API
Side-by-side breakdown.
ai-api
Helicone vs Replicate
Side-by-side breakdown.
Staxly is an independent catalog of developer platforms. Outbound links to Helicone are plain references to their official pages. Pricing is verified at publication time — reconfirm on the vendor site before buying.