How do limits work?
Each plan includes a monthly agent usage budget denominated in dollars. Each Notis message deducts from that budget using Notis model rates, which include a 20% service markup over the provider’s published API price for OpenAI models. You can view your billing cycle, included usage, On-Demand usage, and request breakdowns on the Usage page of your portal. In a Manager thread, the usage indicator also shows the thread’s completed cost and how much context the latest completed run used. If you use Notis with a team, everyone shares one monthly usage pool. For example, a 10-person team on a plan with $150 of included monthly usage per seat gets a shared $1,500 pool for the month. Any teammate can use it, and all team activity counts toward the same total. The Usage page shows the team’s shared usage for the current billing cycle, and billing settings are managed by the team owner. Notis manages model routing automatically based on the task. Prices below are shown per 1M tokens and reflect the current Notis rates after markup.Auto mode
By default, Notis runs on Auto (recommended). Instead of a fixed model, Notis’s smart router reads each turn and picks the intelligence level — model and reasoning effort together — that fits the task: light, inexpensive work for a quick reply, and a stronger model for a hard, multi-step task. Because the choice is made per turn, a single conversation can move up and down levels as the work changes, so you only pay for the capability each step actually needs.Intelligence levels
You can pin an intelligence level — per conversation in the Desktop and Web app, as a per-channel default in Channels, or per automation in the automation editor — instead of letting Auto choose. Each level sets the model and its reasoning effort together:| Level | Model | Reasoning effort | Price vs Medium |
|---|---|---|---|
| High | gpt-5.5 | Medium | ×2 |
| Medium | gpt-5.4 | None | ×1 |
| Low | gpt-5.4-mini | Low | ÷4 |
Processing tiers (priority)
Notis chooses the processing tier automatically, but on a plan with priority access you can pin one per thread or per channel alongside the intelligence level (it shows as Standard / Fast / Economy in the picker):| Tier (picker label) | Price vs Standard | When it’s used automatically |
|---|---|---|
| Fast (Priority) | ×2 | Voice interactions when the model supports it |
| Standard (Normal) | ×1 | Default for regular messages and user-triggered work |
| Economy (Flex) | ÷2 | Background or automation-triggered work when the model supports it |
Automations always run on Economy (Flex) — about half the base price, at a lower (slower) priority. That’s fixed and separate from the per-automation Intelligence setting: pinning an automation to High or Low changes only its model and effort, never its priority, so it keeps the Flex discount either way.
- Normal
- Flex
- Priority
| Model | Input | Cached input | Output |
|---|---|---|---|
| gpt-5.5 | $6.00 | $0.60 | $36.00 |
| gpt-5.4 | $3.00 | $0.30 | $18.00 |
| gpt-5.4-mini | $0.90 | $0.09 | $5.40 |
| gpt-5.4-nano | $0.24 | $0.024 | $1.50 |
| o4-mini-deep-research | $2.40 | $0.60 | $9.60 |
| Tool model | Pricing basis | Notis rate |
|---|---|---|
Nano Banana Pro (gemini-3-pro-image-preview) | Text input / image input / reasoning / text output | $2.40 / $2.40 / $14.40 / $14.40 per 1M tokens |
Nano Banana Pro (gemini-3-pro-image-preview) | Image output | $144.00 per 1M tokens |
| gpt-image-2 | Text input / cached text input / text output | $6.00 / $1.50 / $12.00 per 1M tokens |
| gpt-image-2 | Image input / cached image input / image output | $9.60 / $2.40 / $36.00 per 1M tokens |
| whisper-1 | Transcription | $0.0072 per minute |
| gpt-4o-mini-tts | Text input / audio output | $0.72 / $14.40 per 1M tokens |
| sora-2-pro | 720p video output | $0.36 per second |