Pricing for orchestrating coding agents
Transparent pricing with weighted fast requests for priority runs and a separate monthly token budget for total included compute.
Free
Get started with AI-powered development
50 weighted fast requests/mo
50K standard token budget/month
- 50 weighted fast requests/month
- 50K standard token budget/month
- Cross-device sync
- Mobile app access
- Managed models within fast quota; premium models use higher weights
- No slow queue
- Best for trying the full MiraBridge workflow
Pro
Professional plan for developers
150 weighted fast requests/mo
2M standard token budget/month
- 150 weighted fast requests/month
- 2M standard token budget/month
- Cross-device sync
- Mobile app access
- All managed AI models while fast quota is available.
- Slow queue for lighter models after fast quota
- Built for daily AI-assisted development
- Codex Bridge
Pro+
PopularFor power users and teams
280 weighted fast requests/mo
3M standard token budget/month
- 280 weighted fast requests/month
- 3M standard token budget/month
- Cross-device sync
- Mobile app access
- All managed AI models while fast quota is available.
- Slow queue for lighter models after fast quota
- Power features for heavier usage
- Codex Bridge
Ultra
High-volume AI coding for enterprises
900 weighted fast requests/mo
9M standard token budget/month
- 900 weighted fast requests/month
- 9M standard token budget/month
- Cross-device sync
- Mobile app access
- All managed AI models while fast quota is available.
- Slow queue for lighter models after fast quota
- Maximum throughput and priority access
- Codex Bridge
- Dedicated support
Two budgets, one bill
Fast requests are your enforced priority allowance. The token budget is your standard monthly compute budget for tracked SaaS usage — higher-cost models consume it faster. Claude Opus currently uses 15 fast request credits per priority run. Paid plans continue with lighter models in the slow queue after fast credits are exhausted.
Available AI models
Live model catalog from the MiraBridge Gateway, with public API input and output pricing per 1M tokens.
| Model | Provider | API Price / 1M | Context Window |
|---|---|---|---|
| No published models are available right now. | |||
Need more fast credits?
Purchase extra fast-request credits anytime. Packs never expire.
Slow queue for lighter models
When your fast-request credits run out, paid plans keep working through the slow queue for lighter managed models.
Useful fallback
Lighter managed models remain available for queued work after fast credits run out.
5–30s delay
Requests are queued with a short delay depending on server load. Off-peak is nearly instant.
20 per hour
Rate limited to 20 slow requests per hour. Need more? Buy a request pack for extra fast-request credits.
Compare plans
See exactly what you get with each plan.
| Feature | Free | Pro | Pro+ | Ultra |
|---|---|---|---|---|
| Weighted fast requests/mo | 50 | 150 | 280 | 900 |
| Standard monthly token budget | 50K | 2M | 3M | 9M |
| Managed AI models | Within fast quota; premium models use higher weights | Published models while fast quota is available | Published models while fast quota is available | Published models while fast quota is available |
| Mobile app access | ||||
| Cross-device sync | ||||
| Plan mode | ||||
| Cascade flow | ||||
| Debug mode | ||||
| Brainstorm / TDD / Verification | ||||
| Batch tool approval | ||||
| BYOK (Bring Your Own Key) | ||||
| Priority queue | ||||
| Concurrent sessions | 1 | 2 | 5 | 10 |
| Cloud agents | ||||
| Codex Bridge | ||||
| Dedicated support |
Need custom limits or team controls?
Route provider keys, set team-level limits, and keep agent work behind approval gates.
Frequently Asked Questions
Run your first agent session in under five minutes
Install the editor extension, sign in once, and start orchestrating. Free to try. Bring your own keys if you prefer.
Free to start, no credit card. Bring your own keys to remove limits.