Dollar budgets
Set a cap before a risky run and the agent stops when it hits zero - even mid-step.
Set a cap before a risky run and the ag…Evidence
Billing
Model routing & cost control
Kodus keeps eight providers and dozens of models behind one inbox-style composer. Routing lets you pin cheap models to boring work and flagship models to surgical edits - or drop to a local Llama so the meter stays at zero.
- Monthly subscription: Predictable seat billing with pooled tokens and iterations. Best when the whole squad lives in Kodus daily.
- Bring your own keys: Keep your negotiated rates with OpenAI, Anthropic, Google, or others - Kodus becomes the control plane, not the invoice.
- Pay-as-you-go credits: Freelancers and bursty teams load a balance, earn bonus credits on larger top-ups, and burn them when work appears.
model-routing-cost-control-overview.md - operating view
DollarMonthly subscription
LiveBring your own keys
HeadroomPay-as-you-go credits
PurchasedLocal hardware
Billing
Route by kind of work, not vibes
The Routing tab lets you decide which capability goes to which model family. Simple file reads drift to economical models; multistep refactors inherit the heavyweight you trust for production. Vision-heavy tasks stay on multimodal-capable endpoints.
- Monthly subscription: Predictable seat billing with pooled tokens and iterations. Best when the whole squad lives in Kodus daily.
- Bring your own keys: Keep your negotiated rates with OpenAI, Anthropic, Google, or others - Kodus becomes the control plane, not the invoice.
- Pay-as-you-go credits: Freelancers and bursty teams load a balance, earn bonus credits on larger top-ups, and burn them when work appears.
- Local hardware: Ollama and LM Studio traffic never hits a usage bill; you pay for GPUs once, not per token.
- Dollar budgets: Set a cap before a risky run and the agent stops when it hits zero - even mid-step.
Active focus:
Billing
Scope
Risk
Output
model-routing-cost-control-overview.md
Capturing...
// Focus:
Route by kind of work, not vibes
1) Baseline one squad: Export four weeks of actual token + iteration data before you preach best practices. 2) Publish guardrails: Document default routes, when Review mode is mandatory, and how to escalate spend. 3) Automate warnings: Hook headroom metrics into whatever alerts your FinOps team already reads. 4) Rebalance quarterly: Model leaderboards move fast; reroute the cheap seats whenever the gap closes.
1Dollar
2Live
3Headroom
4Purchased
5Review
What Billing stakeholders get operationally
Live ticker
Balances update while tokens stream so nobody sees surprise invoices weeks later.
Balances update while tokens stream so…Evidence
Headroom meters
Separate windows track iterations across four-hour, weekly, and monthly horizons with reset countdowns.
Separate windows track iterations acros…Evidence
High-signal placements
Billing rollout focus
The Routing tab lets you decide which capability goes to which model family. Simple file reads drift to economical models; multistep refactors inherit the heavyweight you trust for production. Vision-heavy tasks stay on multimodal-capable endpoints.
Dollar budgets
Set a cap before a risky run and the agen…Live ticker
Balances update while tokens stream so no…Headroom meters
Separate windows track iterations across…Purchased credits
Top-ups pile into your wallet without exp…Monthly subscription
Buyer-ready evidence
Pilot
Review
Evidence
Scale
How it works
The Routing tab lets you decide which capability goes to which model fa…
The Routing tab lets you decide which capability goes to which model family. Simple file reads drift to economical models; multistep refactors inherit the heavyweight you trust for production. Vision-heavy tasks stay on multimodal-capable endpoints.
1) Baseline one squad
Export four weeks of actual token + iteration data before you preach best practices.
2) Publish guardrails
Document default routes, when Review mode is mandatory, and how to escalate spend.
3) Automate warnings
Hook headroom metrics into whatever alerts your FinOps team already reads.
FAQ
FAQ
Can I mix subscription and BYO keys?
Yes - pick the combination that matches each workspace. Just make sure finance knows which cost center owns which key.
What happens when a budget trips mid-run?
The agent halts before spending more. You raise the cap or trim the ask, then resume.
Do bonus credits expire?
Purchased credits don’t time out. Subscription allotments still follow your plan’s reset rules.
How do I prove savings?
Compare token or dollar mix per workflow before and after routing rules; report the delta with your finance math, not a universal benchmark.
How should we pilot?
Pick one bottlenecked workflow with named reviewers, run two cadence loops, revisit metrics.
Does tooling replace approvals?
No - Kodus complements review, scanners, budgets, and your escalation paths.
Pricing
Pricing
Use the same Kodus plans, tokens, and routing controls across workflows and posture.
Best to Start
Starter
For individual usage.
$20/mo
- 10M tokens / month
- 300 iterations / month
- Standard model routing
- Chat history + projects
- Community support
Team
For small teams.
$100/mo
- 70M tokens / month
- 2,500 iterations / month
- Full routing + Review + Strategy
- Bring your own local model
- Teams (up to 2 members)
- Priority support
- Audit log access
Scale
For larger organizations.
$200/mo
- 300M tokens / month
- 7,500 iterations / month
- Unlimited team members
- All models + custom routing
- Dedicated support channel
- Early access to beta features
- No annual contract
- Tokens reset monthly
- Switch plans anytime
Have invite code? Get Access Now