Pricing built for indie AI builders
Start free. Pay when you ship. No seat-based nonsense. No overages.
Free
For side projects and prototypes that don't need production reliability yet.
- ✓1 project
- ✓10,000 calls / month
- ✓7-day log retention
- ✓Cost + latency dashboard
- ✓Log inspection
- ✓1 cost alert (email)
- ✕Prompt version tracking
- ✕30-day retention
- ✕Advanced alerts
Builder
For apps in production that need real visibility before something breaks.
- ✓10 projects
- ✓50,000 calls / month
- ✓30-day log retention
- ✓Prompt version tracking
- ✓Side-by-side version comparison
- ✓Replay requests (copy as curl)
- ✓3 cost + latency alerts
- ✓Export logs (CSV / JSON)
Pro
For revenue-generating apps that can't afford a broken prompt in production.
- ✓Unlimited projects
- ✓250,000 calls / month
- ✓90-day log retention
- ✓Version tracking + comparison
- ✓Replay requests
- ✓A/B prompt comparison
- ✓JSON + schema eval checks
- ✓Unlimited alerts
- ✓Export logs (CSV / JSON)
Enterprise
For teams that need unlimited scale, SSO, custom retention, and a dedicated support channel.
- ✓Unlimited projects
- ✓Unlimited calls / month
- ✓Custom log retention
- ✓Everything in Pro
- ✓SSO / SAML
- ✓Dedicated Slack support
- ✓SLA guarantee
All plans are hard-capped — no surprise overage charges. Logging pauses at the limit and resumes on the 1st.
Compare plans
Everything you get, side by side.
| Feature | Free | Builder | Pro | Enterprise |
|---|---|---|---|---|
| Usage | ||||
| Calls / month | 10,000 | 50,000 | 250,000 | Unlimited |
| Overage | — | — | — | — |
| Projects | 1 | 10 | Unlimited | Unlimited |
| Log retention | 7 days | 30 days | 90 days | Custom |
| Observability | ||||
| Cost + latency dashboard | ✓ | ✓ | ✓ | ✓ |
| Log inspection | ✓ | ✓ | ✓ | ✓ |
| Cost alerts | 1 (email) | 3 | Unlimited | Unlimited |
| Export logs (CSV / JSON) | — | ✓ | ✓ | ✓ |
| Prompt management | ||||
| Version tracking | — | ✓ | ✓ | ✓ |
| Side-by-side comparison | — | ✓ | ✓ | ✓ |
| Replay requests (copy as curl) | — | ✓ | ✓ | ✓ |
| A/B prompt comparison | — | — | ✓ | ✓ |
| Evaluation | ||||
| JSON + schema eval checks | — | — | ✓ | ✓ |
| Platform | ||||
| 3-line setup | ✓ | ✓ | ✓ | ✓ |
| SSO / SAML | — | — | — | ✓ |
| Dedicated Slack support | — | — | — | ✓ |
| SLA guarantee | — | — | — | ✓ |
How we compare
| Feature | Stemma | Helicone | LangSmith |
|---|---|---|---|
| Free tier | 10,000 calls | 1,000 requests | No |
| Paid starts at | $19/mo | $20/mo | $39/mo |
| Seat-based pricing | No | No | Yes |
| Setup complexity | 3 lines | Moderate | Complex |
| Designed for | Indie devs | Teams | Enterprise |
| Cost tracking | Yes | Yes | Limited |
| Prompt versioning | Yes | Limited | Yes |
Common questions
What counts as a call?
Every LLM call tracked through the Stemma SDK counts as one call. Dev and production calls both count unless you filter by environment.
What happens if I exceed my call limit?
Logging pauses for the rest of the month — your app keeps working normally, we just stop tracking. Limits reset on the 1st. There are no overages or surprise charges on any plan.
Do you store my API keys?
No. API keys are session-only in the Playground and are never sent to our servers. For production, you use your own keys directly with OpenAI or Anthropic — Stemma only receives the call metadata you explicitly log.
Can I export my data?
Yes. Builder, Pro, and Enterprise plans all include CSV and JSON export from the dashboard.
What LLM providers do you support?
OpenAI, Anthropic, and any OpenAI-compatible API. Additional providers are added based on user requests — open an issue on GitHub to vote.
Do you offer refunds?
Yes, we offer a 14-day refund if Stemma doesn't work for your use case. No questions asked.