QP
Cost Centeragent-as-actor · tokens & dollars · usage-based
FinOps · Cost Center
What every AI agent costs — by model, agent, and user.
$684.50 incl. est.
total spend
all agents · All
$606.50
provider-reported
billed by the provider
$78.00
estimated
derived from tokens
54%
from cache
cache-read tokens
Spend by agent
$684.50 totalClaude Code
$214.50
support-copilot
$168.40
Cursor
$96.20
Codex CLI
$78.00
data-pipeline-agent
$52.30
marketing-gpt
$44.10
Claude Desktop
$31.00
Spend over time
daily · last 7d$88.2005-28
$102.5005-29
$96.1005-30
$71.0005-31
$84.4006-01
$119.7006-02
$142.6006-03
By model
5 modelsModelTokensResponsesCost
claude-opus-4-8 7.7M1.4k$245.50
gpt-4o 6.7M4.0k$212.50
claude-sonnet-4-6 3.1M980$96.20
gpt-5-codex est2.4M540$78.00
claude-haiku-4-5 9.2M5.4k$52.30
By agent
7 agentsAgentTokensResponsesCost
Claude Code 6.9M1.2k$214.50
support-copilot 5.6M3.1k$168.40
Cursor 3.1M980$96.20
Codex CLI est2.4M540$78.00
data-pipeline-agent 9.2M5.4k$52.30
marketing-gpt 1.1M860$44.10
Claude Desktop 800.0k180$31.00
By user
6 identitiesUserTokensResponsesCost
mark@qpoint.io 7.7M1.4k$245.50
svc-support@qpoint.io 5.6M3.1k$168.40
jane@qpoint.io 3.1M980$96.20
dev-7f3@duck.com est2.4M540$78.00
svc-data@qpoint.io 9.2M5.4k$52.30
(proxy-only, no identity) 1.1M860$44.10
est rows are estimated from token counts — some providers (OpenAI / Codex) don't report a billed cost, so we derive it from usage. This is usage-based observation, not a billed invoice.
Fields: costs.totals (total_cost · provider_cost · estimated_cost · cache_read_pct) · costs.by_model · by_agent · by_user · series — from app/data/c16-fleet.ts (qcontrol /api/costs shape).