Improve the business outcomes of your AI-powered workflows.
Start with one workflow. Scale across teams with predictable pricing and included AI credits.
Editions
Your plan includes platform access, usage allowances, and AI credits. Add capacity only when you choose.
| Starter Explore DataFramer. $0 platform fee No credit card. Start free | Business Operate team workflows. $999 / month Shared capacity across projects and agents. Get started | Enterprise Scale across departments. Custom Annual pricing Scope your rollout | |
|---|---|---|---|
| Deployment | DataFramer Cloud | DataFramer Cloud | VPC or DataFramer Cloud |
| Support | Self-service | Standard support (Email, Slack, Video) | Priority support with SLAs |
| Advisory & Implementation | — | — | Add-on |
| AI spend (BYOM or Credits) | $5 credits included. $20 / month limit. | $200 credits included. $1000 / month limit. | Included annual credit allowance. |
| Core platform capabilities | Included | Included | Included |
- Enterprise offers optional VPC deployment, contracted support, and
expandable capacity. With VPC deployment, your data plane runs in your cloud account; model
calls follow your configured provider and deployment settings.
- BYOM = Bring your own model credential: an Anthropic or OpenAI API key, or a Databricks service principal (Enterprise). Credits pay for model calls through DataFramer at provider rates. With your own credential, you pay your provider directly; DF credits do not apply.
Limits
| Starter | Business | Enterprise — tailored capacity | |
|---|---|---|---|
| Users | 5 | 25 | Custom |
| Monthly ingestion | 1 GB | 50 GB | Custom |
| Retention | 14 days | 60 days | Custom |
How usage and credits work
Ingestion is pooled across your workspace, covers events and traces, and is measured before compression, excluding internal DataFramer-generated records. Business credits expire monthly, including on annual plans. Enterprise credits expire annually. Starter credits are one-time.
How very large traces are processed
A trace too big for a single model call is split into windows and analyzed across them, not dropped. A unit is only set aside when it still cannot be analyzed — a single indivisible piece over the per-call budget, or an attachment we cannot read, which we skip rather than analyze as degraded content. Every one of those is recorded with its reason and counted in the run, so coverage is always visible rather than silently reduced. On the evaluations API the same condition returns an explicit error instead of a skip.
Overages and options
Additional AI usage
Prepaid DF credits or bring your own model credential (BYOM).
Additional ingestion
$3 per additional GB.
Extended retention
$0.50 per retained GB / month beyond the plan's included retention.
Business users
Additional seats can be purchased.
Enterprise expansion
Higher allowances and longer retention at contracted pricing.
VPC
VPC split deployment is scoped to your environment. Customer cloud costs are separate.
Separately scoped
Custom integrations and additional environments.
Education & Implementation
Establish the operating model for an important AI workflow, then implement it with DataFramer, or take the model and run it yourself.
Implementation
$7,500+ one-time
Implement one important workflow with clear outcomes, quality standards, measurement, and review.
Advisory + Implementation
$15,000+ one-time
Define the operating model for one workflow, align business and technical teams, and run the first improvement cycle.
No card · All core product capabilities included