Track OpenAI Token Usage from Your Mac Menu Bar
ChatGPT Plus caps GPT-4o at ~40 messages per 3-hour window. API bills accumulate silently. AIUsageBar shows both in real time so you stay in control.
How OpenAI usage limits work
- ChatGPT Plus: ~40 GPT-4o messages per 3-hour rolling window, then falls back to GPT-4o mini.
- OpenAI API: token-based billing — GPT-4o costs $2.50/1M input, $10/1M output tokens.
- Monthly API spend accumulates silently; the dashboard updates with up to 1-hour delay.
What AIUsageBar shows you
- ChatGPT Plus message count and window reset timer
- API token usage by model (GPT-4o, o3, o4-mini)
- Estimated daily and monthly API cost
- Spend alerts before you hit your budget cap
What is OpenAI?
OpenAI is the company behind the GPT model family and ChatGPT — the consumer AI product that brought large language models to mainstream adoption. Their model lineup in 2026 includes GPT-4o (fast, multimodal), o3 and o4-mini (reasoning-focused), and GPT-4.5. OpenAI operates two distinct products with very different billing and limit models: ChatGPT (a subscription product for consumers) and the OpenAI API (a token-metered developer platform).
OpenAI's models are also embedded in many third-party developer tools. Cursor uses GPT-4o as one of its primary models. GitHub Copilot is powered by OpenAI models. Understanding your total OpenAI exposure means accounting for all of these surfaces, not just what you use directly in ChatGPT.
OpenAI Rate Limits and Costs: A Full Breakdown
ChatGPT Plus ($20/month)
ChatGPT Plus limits GPT-4o access to roughly 40 messages per 3-hour rolling window. After that, ChatGPT suggests switching to GPT-4o mini (which is uncapped). The o3 model has its own separate and lower message limit. The 3-hour window resets from the time of your first message in the period — like Claude, it's rolling, not fixed.
ChatGPT Pro ($200/month)
Pro removes most rate limits on GPT-4o and includes access to o1 Pro mode. For developers who rely on ChatGPT as a daily driver, this tier is primarily for removing friction rather than for API access.
OpenAI API
API usage is billed per token. Pricing as of 2026:
- GPT-4o: $2.50/1M input tokens, $10/1M output tokens
- GPT-4o mini: $0.15/1M input tokens, $0.60/1M output tokens
- o3: $10/1M input tokens, $40/1M output tokens
- o4-mini: $1.10/1M input tokens, $4.40/1M output tokens
API rate limits are set per organization by tier (1–5) and expressed as RPM and TPM per model. New organizations start at Tier 1 with conservative limits; tiers increase with cumulative spend history. You can also set a monthly hard spend limit in the API dashboard.
Why Tracking OpenAI Usage Matters
OpenAI's usage dashboard at platform.openai.com/usage updates with up to an hour of delay. Email alerts fire after you've already hit a threshold — not before. If you're running automated workflows, building with the API, and also using ChatGPT manually, your total OpenAI exposure is scattered across multiple surfaces with no single ambient view.
For developers using Cursor (which uses GPT-4o), the situation is further obscured: your Cursor Pro fast requests drawing from GPT-4o don't appear in your personal OpenAI usage dashboard — they're billed to Cursor. But Cursor's own request counter doesn't show cost in dollars. Without a unified tracker, your real AI spend is invisible.
A $200 month of API usage is easy to build up in 3 weeks of heavy development, especially when o3 is in the mix. Real-time spend visibility is the difference between managing your budget and getting surprised by it.
How AIUsageBar Tracks OpenAI
AIUsageBar shows ChatGPT Plus message count and the remaining time in your 3-hour window — the exact data the ChatGPT interface doesn't surface prominently. For API users, it tracks live token spend by model and projects your monthly cost based on the current daily burn rate. You can set a daily or monthly budget threshold and receive a notification before you hit it, not after.
Frequently Asked Questions
Does ChatGPT Plus have a message limit on GPT-4o?
Yes — roughly 40 messages per 3-hour rolling window. After that, ChatGPT automatically suggests GPT-4o mini. The o3 model has a separate and lower limit of roughly 50 messages per week on Plus.
How does OpenAI API billing work?
You're charged per token — separately for input (the prompt you send) and output (the response you receive). Costs vary by model. Your bill accumulates through the month and is charged to your payment method on the billing date. You can set a monthly hard cap at platform.openai.com/account/limits.
What's the difference between ChatGPT Plus and the OpenAI API?
ChatGPT Plus is a $20/month subscription to the ChatGPT interface. The OpenAI API is a separate, usage-metered service for developers building on top of OpenAI's models. They have separate billing, separate rate limits, and separate dashboards. Many developers use both.
Can I see my total OpenAI spend across ChatGPT and the API?
Not natively — they're billed separately. AIUsageBar consolidates both into a single view alongside your other AI tools, so you always know your real total spend at a glance.