Every dollar traced. Every angle covered.
Your LLM provider sends a monthly total. Spaces captures cost at the token level and tags it with full context — task, workflow step, model, user, plan. Then lets you roll it up along any axis to manage your operations with real data, not estimates.
Cost captured at the token level, attributed to every task as work happens — not reconstructed from a monthly invoice.
Automatic cost intelligence, built into the workflow
No SDKs to wrap around API calls. No billing logs to parse. Spaces captures token usage directly from your coding sessions and tags every cost with the full context of the work that triggered it.
Capture
Every coding session reports token consumption automatically. Input tokens, output tokens, model, cost — captured at the source from your actual development environment. Not reconstructed from invoices.
Tag
Each usage event carries full context: which task, which workflow step, which user, which plan, which model. Cost isn't "per workspace" or "per day." It's per unit of work, with every dimension attached.
Roll up
Aggregate the same granular data along any axis. By plan, by user, by workflow, by model — and combinations of these. Same underlying data, as many perspectives as your operations need.
Roll up along any axis. Answer any question.
The same granular cost data, surfaced through whatever lens your operations need. Because every cost record carries its full context, you choose the dimension — not the other way around.
By project plan
Roll up every task in a plan to see what it actually cost — not what you estimated. Compare plans side by side to understand where your budget goes and which types of work are most expensive.
By workflow
Users define the workflows that structure their work. See what each workflow costs end-to-end across every task that ran it — then drill into any workflow to see cost by step.
By user
See how costs distribute across your team. Spot whether expensive tasks cluster around specific people, identify who might benefit from better tooling, and understand how team structure affects total spend.
By model
Compare model costs on real workloads, not benchmarks. See which models are used where, whether expensive models are doing work that cheaper ones could handle, and how model choices affect total spend.
Cross-dimensional analysis
Roll up costs however you need, and segment them too. See costs by workflow, then segment by step within each one. See costs by plan, then segment by model. When a plan is sourced from a document, cost data traces all the way back — giving you a real measure of how well your documentation prepared the work. The combinations reveal what flat summaries never can.
Beyond visibility — control and optimization
Tracking costs is the foundation. Rolling them up along the right axes is where the value compounds.
Budget guardrails
Set spending limits at any level — task, plan, user, or model. Alerts trigger before thresholds are breached, when a budget is approaching its limit rather than after it's exceeded.
Cross-plan comparison
Compare actual cost profiles across plans to spot efficiency patterns and outliers. Understand which types of plans are consistently expensive and which deliver the most output per dollar.
Spend forecasting
Historical cost-per-task-type profiles feed directly into the projection engine. Forecast remaining spend for in-progress plans based on actual data, not estimates.
Operational insights
Combine dimensions to surface patterns invisible in flat summaries. Costs by step within a plan, by model within a user — the cross-references reveal where your real optimization opportunities are.
Stop guessing what agents cost. Start seeing the full picture.
Every token attributed to the task that consumed it. Every cost rolled up by plan, user, workflow, and model. Every budget tracked before it's exceeded. This is cost intelligence built into the workflow.
Free during beta · No credit card required