Cursor usage dashboard
The Cursor usage dashboard at cursor.com/dashboard is billing truth for your cycle: spent, remaining allowance, on-demand charges, and token breakdown by model.
This page answers searches for cursor.com/dashboard usage, remaining allowance, usage-based pricing, and spending limits.
Hub: Tokenminning in Cursor.
Open the dashboard
- Go to cursor.com/dashboard
- Open Usage for the current billing cycle
- Open Billing for subscription, usage-based toggle, and spending limit
Usage tab fields
| Field | Meaning |
|---|---|
| Spent | Included allowance used so far (USD) |
| Remaining allowance | Left before exhaustion or on-demand billing |
| On-demand charges | Pay-as-you-go after included usage (if enabled) |
| Token breakdown | Input, cache write, cache read, output by model |
| Reset date | Cycle rollover — unused allowance does not carry forward |
Before a long Agent session, check remaining allowance. If the API pool is low, switch to Auto or Composer so work draws from the First-party models pool.
Two usage pools
Individual plans track two pools that reset monthly:
| Pool | Models | Allowance |
|---|---|---|
| API | Named models you pick — Sonnet, GPT, Opus, etc. | Monthly USD; scales with plan — Usage and limits |
| First-party models | Auto, Composer 2.5, Grok 4.5 | Separate, more generous meter |
Frontier models burn the API pool fastest. Auto and Composer use the First-party pool — default them for routine agent work.
Queries about Auto + Composer vs API usage pool breakdown refer to this split. Check both meters in the dashboard after a heavy week.
Billing tab
cursor.com/dashboard → Billing:
| Setting | Purpose |
|---|---|
| Subscription tier | Plan and renewal |
| Usage-based pricing | Continue at API rates after included usage runs out |
| Spending limit | Hard cap on on-demand charges |
| Manage Subscription | Plan changes and usage reset time |
Tab completions are unlimited on individual plans — you meter Agent, Chat, and named-model API calls.
In-editor usage indicator
Quick remaining allowance snapshot while coding. Use mid-session; reconcile against the dashboard after heavy weeks.
Teams
- Analytics API / Admin API
- Teams dashboard for per-user usage
- Non-Auto third-party models: API rates plus Cursor Token Rate
Debug from token groups
After a heavy Agent week:
| Dominant group | Likely fix |
|---|---|
| Input | Shorten rules; disable MCP |
| Cache read | New chat per task |
| Cache write | Narrow attachments |
| Output | Auto/Composer; tighter prompts |
Guardrails
- Set a spending limit in Billing before enabling usage-based pricing
- Know which pool you are drawing from
- Team admins: analytics in Teams dashboard