Limits & Plans
Understanding your usage limits and plans.
Plan Comparison
| Feature | Free | Pro | Max |
|---|---|---|---|
| Daily tokens | 15,000 | 150,000 | 300,000 |
| AI models | 11 | 11 | 11 |
| Max output tokens | 750 | 1,000 | 1,250 |
| Max document input | 8,000 | 12,000 | 15,000 |
| Custom presets | 3 | 100 | 100 |
| Folders | 30 | 100 | 100 |
| Research projects | 3 | 100 | 100 |
| Sources per project | 3 | 100 | 100 |
| Premium TTS | — | 30,000 chars/day | 100,000 chars/day |
| Auto-highlights / day | 3 | 100 | 23 |
| PDF/DOCX | ✅ | ✅ | ✅ |
Chat Limits
| Feature | Free | Pro | Max |
|---|---|---|---|
| Messages per summary | 5 | 100 | 20 |
| Max output tokens | 300 | 1,200 | 1,600 |
| Max context tokens | 2,000 | 6,000 | 8,000 |
Rate Limits
| Action | Free | Pro | Max |
|---|---|---|---|
| Summarize / hour | 10 | 100 | 300 |
| Video / hour | 5 | 20 | 150 |
| Chat / minute | 5 | 10 | 10 |
| API calls / minute | 2 | 20 | 300 |
How Limits Work
5minread uses a token-based system for text and characters for TTS.
Daily Reset
- All limits reset every 24 hours
- Check reset time in the extension popup
Cached Content
- Previously summarized pages are cached
- Reopening cached content doesn't use tokens
Token Limits
What Are Tokens?
Tokens are units of text that the AI processes:
- ~1 token = ~4 characters in English
- ~1 page = ~500–1,000 tokens
- Longer pages use more tokens
Daily Token Allowance
| Plan | Daily Tokens |
|---|---|
| Free | 15,000 |
| Pro | 150,000 |
| Max | 300,000 |
Daily tokens cover all AI operations: summarization, chat, explain, and auto-highlights.
Output Token Limits
Each summary has a maximum output size based on your plan:
| Plan | Max Output Tokens | ~Words |
|---|---|---|
| Free | 750 | ~300 |
| Pro | 1,000 | ~400 |
| Max | 1,250 | ~500 |
Input Token Limits
The amount of source content the AI can process per request:
| Feature | Free | Pro | Max |
|---|---|---|---|
| Document (page/PDF) | 8,000 | 12,000 | 15,000 |
| Video chunk | 4,000 | 4,000 | 4,000 |
| Video final summary | 6,000 | 10,000 | 15,000 |
| Explain | 2,000 | 3,000 | 4,000 |
| Chat | 1,200 | 4,000 | 6,000 |
Auto-highlights
Auto-highlights run as a separate AI pass and count against your daily tokens. Daily limit per plan:
- Free: 3 auto-highlights / day
- Pro: 100 auto-highlights / day
- Max: 23 auto-highlights / day
File Attachments (PDF / DOCX)
Maximum upload size per file:
- Free: 3 MB
- Pro: 10 MB
- Max: 15 MB
What Uses Tokens
- Summarizing new pages
- Chat conversations
- Explain requests
- Auto-highlights
- Custom instructions add minimal overhead
What Doesn't Use Tokens
- Opening cached summaries
- Adjusting settings
- Browsing history
- Exporting summaries
TTS Limits
Premium TTS voices are available on Pro and Max plans.
TTS Types
Standard TTS (All Plans)
- Uses browser's built-in voices
- Unlimited usage
- Quality varies by browser/device
Premium TTS (Pro / Max)
- Neural network voices
- Natural, human-like speech
- Uses your daily character allowance
Plan Limits
| Plan | Premium TTS | Daily Characters | Rate Limit |
|---|---|---|---|
| Free | No | 0 | — |
| Pro | Yes | 30,000 | 20/hour |
| Max | Yes | 100,000 | 200/hour |
~1,000 characters ≈ 150 words ≈ 1 minute of speech. Pro gets ~30 minutes/day and Max gets ~100 minutes/day of premium TTS.
How Characters Are Calculated
- Based on text length, not playback time
- Speed settings don't affect character usage
Audio Export (MP3)
Exporting summaries as MP3 uses your premium TTS allowance. Once exported, you can listen to the file unlimited times offline.
AI Models
Available models depend on your plan. Change the active model in Settings → AI Model.
| Model | Multiplier | FREE | PRO | MAX |
|---|---|---|---|---|
| Mistral Nemo (default) | 1x | ✅ | ✅ | ✅ |
| Meta Llama 3.1 | 1x | ✅ | ✅ | ✅ |
| Ministral | 1x | ✅ | ✅ | ✅ |
| OpenAI o4-mini | 2x | 1/day | 100/day | 100/day |
| Google Gemini 2.0 | 2x | 1/day | 100/day | 100/day |
| Qwen 2.5 | 2x | 1/day | 100/day | 100/day |
| Qwen 3 | 2x | 1/day | 100/day | 100/day |
| OpenAI GPT-5 | 2x | 1/day | 100/day | 100/day |
| Meta Llama 3.3 | 3x | 1/day | 100/day | 100/day |
| Google Gemini 2.5 | 2x | 1/day | 100/day | 100/day |
| DeepSeek R1 Distill | 2x | 1/day | 100/day | 100/day |
All 11 models are available on every plan. The difference is the token multiplier and a daily cap on expensive models for Free users.
Standard models — 1x multiplier (unlimited on all plans)
- Mistral Nemo (default), Meta Llama 3.1, Ministral
- Fast, efficient, good quality for everyday content
Expensive models — 2–3x multiplier
- o4-mini, Gemini 2.0, Qwen 2.5, Qwen 3, GPT-5, Llama 3.3, Gemini 2.5, DeepSeek R1
- Higher quality, deeper reasoning — but cost 2–3× more tokens per request
- Free: limited to 1 use/day per expensive model
- Pro / Max: 100 uses/day (effectively unlimited)
Checking Your Usage
The extension popup shows your current usage:
- Token percentage used
- TTS characters remaining (Pro/Max)
- Time until daily reset
View detailed usage in your profile or dashboard on the web app.
When Limits Are Reached
When you reach a limit:
- A notification appears
- Shows time until reset
- Option to upgrade plan
Optimization Tips
Tokens
- Use Short summary length to consume fewer tokens
- Reopen cached summaries — they're free
- Standard preset is well-balanced
- Avoid Max-tier models on Pro if token efficiency matters (3x multiplier)
TTS
- Standard TTS is unlimited — use it freely on any plan
- Export summaries as MP3 for offline replay without using more characters
- Use shorter summary lengths to reduce character consumption
Upgrading
To get more limits:
- Click Upgrade in the limit notification
- Or go to your Profile → Plans
- Choose Pro or Max plan