Pricing
Start free. Scale as your agents scale. No per-token fees, ever.
Free
For personal projects and small experiments.
Billed monthly
Get Started- 500 agent runs / month
- Core observability
- 7-day data retention
- 1 seat
Developer
For solo developers shipping agents in production.
Billed monthly
Start trial7-day free trial
- 2,500 agent runs / month
- Core observability
- Search + filters
- User feedback
- Local-first mode
- 14-day data retention
- 1 seat
Pro
For teams running scale-out workloads with higher retention.
Billed monthly
Start trial7-day free trial
Everything in Developer, plus:
- 25,000 agent runs / month
- Silent failure detection
- Topic clustering
- 60-day data retention
- Up to 5 seats
- Slack support
Enterprise
For regulated deployments and dedicated infrastructure.
Everything in Pro, plus:
- Unlimited seats
- Custom rate limits
- Custom data retention
- SSO login
- SLA guarantees
- Prioritized support
Compare features
Everything in one platform — observability, replay, and recovery.
Frequently asked questions
An agent run is a single top-level invocation of an agent — every LLM call, tool invocation, and checkpoint inside it is included. We never charge per span or per token.
Yes. You can upgrade or downgrade at any time. Annual plans are prorated on upgrade; downgrades take effect at the next billing cycle.
The Free plan is permanently free up to 500 agent runs / month. For early-stage startups above that limit, reach out to us — we run a credits program.
Yes, on the Enterprise plan. We deploy into your AWS / GCP / Azure account; no data ever leaves your infrastructure.
Annual plans are billed once per year and save roughly 20% vs. monthly. You can cancel at any time and keep access through the end of the paid period.
Yes. If you maintain an open-source agent framework or are enrolled as a student, email hello@omium.ai for a complimentary Developer plan.