How Usage & Credits Work
How Apogee meters AI usage - cost-based credits, your monthly allowance, the $50 signup bonus, and how to read your usage in dollars and percent. Starter through Teams & Enterprise.
How Usage & Credits Work
Apogee meters usage by the actual cost of the AI work each request does, measured in dollars. This page explains how that credit model works, what affects your cost, and how to read your usage.
How Usage Is Measured
Every time you interact with Apogee - through the chat client, via MCP, or when a scheduled briefing runs - an AI model processes your request. Usage is metered by cost, not by a fixed number of requests. Two things drive the cost of a request:
- How much text the model processes - both the input (your question plus the context and sources pulled in) and the output (the response). Longer, deeper research costs more than a quick lookup.
- Which model runs - more capable models cost more per unit of text. Apogee selects the model for the task; there is no preset to choose.
Cost is tracked in dollars and shown as a percentage of your plan allowance. There is nothing to denominate in tokens on your side - the interface shows a percent used and the dollar figures behind it.
What Counts as Usage
Different activities cost different amounts:
- Short factual lookups (e.g., "What's the status of HR 1234?") cost a few cents.
- Detailed analysis (e.g., "Compare these two bills and summarize the differences") costs more, because the model reads and reasons over more text.
- Research queries that search across multiple data sources cost the most - the research agent may call several internal tools in sequence, each adding to the total.
- Follow-up questions in the same conversation include earlier messages as context, so a long conversation gradually costs more per message.
Practical example: A quick question like "Who sponsors HR 5678?" costs only a few cents. A detailed research request like "Find all AI regulation bills this Congress and summarize their approaches" costs more, because the agent searches multiple sources and reasons over far more text.
Where Usage Comes From
Your allowance is shared across every way you use Apogee. Each source draws from the same monthly credit balance.
Chat Client
When you send a message in the chat client, the model reads your message plus the conversation history and generates a response. If the research agent calls tools while answering, those internal calls are part of the same request's cost.
MCP Connections
If you connect via MCP (through Claude Desktop, Claude Code, Cursor, or another MCP client), tool calls and AI processing draw from the same balance as the chat client. There is no separate allocation - all usage is pooled.
Briefings
Briefings (scheduled monitoring) draw from your balance when they run. Each run triggers a research query on your behalf to search sources and generate a summary. Briefings run on a schedule and use the same credits as the rest of your work.
If your balance is exhausted when a briefing is scheduled to run, that run is skipped and you are notified. It attempts to run again at the next scheduled time.
Your Credit Allowance
Each plan includes a monthly credit allowance measured in dollars of AI cost. The allowance resets at the start of each billing cycle. Unused credit from one cycle does not carry into the next; the balance refreshes to the plan allowance.
The $50 signup bonus
Every new account starts with a $50 signup bonus in credit. The bonus never expires - it sits alongside your plan allowance and is there whenever you need it, whether that is your first week or several months in. It is the simplest way to run real research before deciding on a plan.
Soft enforcement
Apogee uses soft enforcement: normal usage is never blocked mid-task. Costs are recorded against your balance as you work, and the interface shows how much of your allowance you have used. A daily safety ceiling guards against a runaway loop, but it sits well above normal professional use and does not affect day-to-day work.
Reading Your Usage
You can see your usage in two places:
- Chat header - A usage indicator appears once you have used a meaningful share of your allowance. It shows your percentage used and turns red as you approach the limit.
- Billing page - Visit chat.apog.ai/billing for your current plan, credit balance, dollar usage, and upgrade options.
Plans
All plans include every intelligence capability and every data source. Plans differ by monthly credit allowance and team features.
| Plan | Who it is for | Monthly credit |
|---|---|---|
| Starter (free to start, then $50/month) | Individuals getting started | Entry allowance |
| Pro (from $150/month) | Daily professional use | Larger allowance; rate set automatically by organization type (nonprofit, government, or commercial) |
| Teams & Enterprise (custom) | Teams and organizations | Shared allowance sized to the organization, plus collaboration features |
See pricing for current numbers.
What Happens When You Reach Your Limit
When you have used your plan's credit allowance for the cycle:
- You see a usage indicator in the chat header and a note above the chat input.
- Your existing conversations and data are unaffected - you can still browse previous chats.
- Your allowance refreshes at the start of your next billing cycle.
- You can upgrade for a higher allowance from the Billing page.
Tips for Getting More From Your Credits
Be specific in your questions
Vague questions like "Tell me about healthcare policy" make the agent search broadly and write long answers. Specific questions like "What committees is the BIOSECURE Act assigned to?" cost far less.
Reuse what is already in context
When the agent retrieves data (bill status, news articles), that data stays in the conversation. Follow-up questions about the same data cost less than starting a fresh search.
Start a new conversation for a new topic
Each message carries the earlier conversation as context, so a very long thread costs more per message. Starting a new conversation for an unrelated question resets that context.
Frequently Asked Questions
Does MCP usage count against my allowance?
Yes. Whether you use the hosted chat client or connect via MCP (Claude Desktop, Claude Code, Cursor, and so on), tool calls and AI processing draw from the same balance.
Do data lookups cost anything?
Simple data retrieval (searching bills, fetching news) costs very little. Most of the cost comes from AI processing - generating analysis, summaries, and research synthesis.
What happens to unused credit?
Plan credit refreshes at the start of each billing cycle and does not roll over. The $50 signup bonus is the exception - it never expires.
Do briefings count against my allowance?
Yes. Each briefing run triggers a research query that draws from your balance, the same as a query you run yourself. If your balance is exhausted when a briefing is scheduled, it is skipped and you are notified.
Why does a long conversation cost more over time?
Each message includes the prior messages as context so the response stays coherent. As the conversation grows, each new message sends more text. Starting a new conversation resets this.
Can I see exactly what I used?
The billing page shows your balance and dollar usage. Per-request breakdowns are not currently exposed in the interface, but usage is tracked and reflected in your percentage as you work.