Severon uses usage limits to keep the service reliable and make capacity available across plans.
Depending on your plan and the features you use, the Usage page may show separate limits for:
- Your current session
- Total weekly usage across models
- Specific models or higher-compute modes
- Voice usage
- Developer or API usage
These limits are not necessarily interchangeable. Reaching one limit may temporarily restrict only that model or feature while the rest of Severon remains available.
Check your current usage
Open Severon and select **Usage** from the account or navigation menu. The page shows the percentage used and, when available, when that allowance resets.
The information displayed in your account is the most accurate source for your current plan. Limits can vary by plan and may change as Severon improves capacity.
If the displayed usage looks wrong
Refresh the page and make sure you are signed in to the correct account. If the value still appears incorrect, contact Support and include:
- Your plan
- The limit or feature affected
- The approximate time the issue occurred
- A screenshot of the Usage page
Do not send passwords, payment-card numbers, or API secrets.
---
What happens when I reach a usage limit?
When you reach a usage limit, Severon may temporarily pause access to the affected model or feature until the allowance resets.
You may still be able to:
- Use another available model
- Use a lower-compute mode
- Continue using features with a separate allowance
- Wait for the displayed reset time
- Upgrade if a higher plan is available to your account
The message shown in Severon should identify the affected limit and, when available, the reset time.
Usage credits
If your account supports usage credits, eligible usage may draw from those credits after the included plan allowance is exhausted. Check the Usage or Billing area for your current balance and availability.
Usage credits do not necessarily remove every safety, capacity, model, or rate restriction.
Still blocked after the reset?
Refresh Severon, sign out and back in, and check the Severon status page. If the limit does not clear, contact Support with a screenshot of the message and the time it appeared.
---
Severon API rate limits
API rate limits control how many requests or how much model usage can be processed over a period of time. They protect reliability and can vary by model, account, plan, and current capacity.
Your current Console, developer documentation, and API responses are the source of truth for the limits applied to your account. Do not rely on limits copied from another account or an older article.
Limits may be measured using one or more of the following:
- Requests per minute
- Input or output tokens per minute
- Concurrent requests
- Daily or monthly usage
- Model-specific capacity
When a request is rate limited
A rate-limited request normally returns an HTTP `429` response. Your application should wait before trying again rather than immediately repeating the request.
Use exponential backoff with a small random delay, and follow a `Retry-After` header when one is provided. Avoid retrying invalid requests, because retries will not correct authentication, billing, or input errors.
Need more capacity?
Check the Severon Console for the limits and upgrade options available to your account. Organizations that need sustained higher capacity can contact Severon Support.
Never include a full API key when asking for help.
---
Fixing “Too many requests” or HTTP 429 errors
An HTTP `429` response means your application sent more work than the account or model can accept at that moment.
Recommended steps
1. Stop immediate automatic retries.
2. Read the response body and any `Retry-After` or rate-limit headers.
3. Wait for the indicated period.
4. Retry with exponential backoff and random jitter.
5. Reduce request concurrency or place requests in a queue.
6. Confirm that your application is not accidentally sending the same request more than once.
If only one model is affected, use another available model only when it is appropriate for the task. Do not continuously switch models to bypass safeguards or account restrictions.
If the error continues
Check the Severon status page and the limits shown in the Console. When contacting Support, include:
- The model and API endpoint
- The HTTP status code
- The request time and time zone
- The request ID, if one was returned
Remove prompts, personal data, credentials, and API keys unless Support specifically requests a safe diagnostic sample.
---
How usage is counted
Severon usage can depend on more than the number of messages you send.
Depending on the feature, usage may include:
- The text and instructions you submit
- Previous conversation context used for the response
- Files or other material included in the task
- The model's generated output
- Tool calls, research, code execution, or other extended work
- The selected model and reasoning level
Long conversations, large files, higher-compute modes, and agentic tasks can therefore use more capacity than a short question.
The Usage page shows the account's recorded usage. Estimates shown before or during a task are informational and may differ from final recorded usage.
To reduce usage, start a new conversation when old context is no longer needed, attach only relevant material, and use the model or reasoning level appropriate for the task.
---
Plan and feature availability
Severon features can vary across Free, Pro, Max, organization, and developer offerings.
A feature may also depend on:
- Your country or region
- The model you selected
- Current service capacity
- A staged product rollout
- Organization administrator settings
- Account eligibility or safety restrictions
The controls visible in your signed-in Severon account show what is currently available to you.
If a feature shown in an announcement is missing, refresh the app, confirm that you are using the correct account, and check your plan. Some features are released gradually and may not be available to every user immediately.
Support can help investigate missing access, but cannot guarantee early access or override every eligibility restriction.
---
Choosing a Severon model
The best model depends on the kind of work you are doing. Models and availability may change, so use the descriptions shown in the model selector as the current source of truth.
As a general guide:
- Use a fast, lightweight model for short questions, drafting, and routine tasks.
- Use a general model for everyday analysis and mixed work.
- Use a reasoning-focused model for difficult planning, math, or multi-step problems.
- Use Severon's most capable model for demanding work where quality matters more than speed.
Higher-capability models or higher reasoning settings can take longer and may consume more of your allowance.
If a model is unavailable, it may be restricted by your plan, temporarily at capacity, retired, or unavailable in your region. Select another suitable model or check again later.
---
Voice mode availability and troubleshooting
Voice mode lets eligible users speak with Severon using a supported browser and microphone.
Voice availability may depend on your plan, browser, device, region, and microphone permissions. The voice control shown in your account indicates whether the feature is currently available to you.
If voice mode does not start
1. Confirm that your plan includes voice access.
2. Allow microphone access when your browser asks.
3. Check the site's microphone permission in browser settings.
4. Make sure another application is not exclusively using the microphone.
5. Refresh Severon and try again.
6. Test with a current version of a supported browser.
Voice usage may have a separate allowance from text usage.
If the problem continues, tell Support your browser, operating system, and whether the microphone works on another site. Do not send private recordings unless you intentionally choose to share them.
---
Signing in to Severon
Severon may use a secure sign-in link or an available identity provider to authenticate your account.
If your sign-in email does not arrive
1. Confirm that the email address was entered correctly.
2. Check Spam, Junk, Promotions, and filtered folders.
3. Wait a few minutes before requesting another link.
4. Use the newest sign-in email, because an older link may have expired.
5. Ask your school or organization administrator whether automated mail is blocked.
Open the sign-in link only on a device and browser you trust. Never send the link, password, or verification code to Support.
If the link is expired, request a new one. If you are repeatedly returned to the sign-in page, allow required cookies for Severon, disable extensions that block authentication, and try again.
---
Canceling a Severon subscription
You can request cancellation from the Billing area of your Severon account when self-service cancellation is available.
Cancellation normally stops automatic renewal. Unless the Billing page says otherwise, paid access continues until the end of the current billing period and the account then moves to the available Free plan.
Before canceling
Review the date shown in Billing and save any invoices you need. Canceling a subscription is different from requesting a refund and does not automatically create refund eligibility.
If the cancellation control is unavailable:
- Confirm that you are signed in to the account that owns the subscription.
- Ask your organization administrator if the plan is centrally managed.
- Contact Support before the renewal date.
Keep the confirmation message or email for your records.
---
Billing, charges, and invoices
Use the Billing area of your Severon account to review your current plan, renewal information, available payment details, and invoices.
If you do not recognize a charge, first confirm:
- The account email associated with the subscription
- Whether a free trial converted to a paid plan
- Whether the plan renewed automatically
- Whether another authorized person purchased the plan
- Whether the charge is pending or completed
Contact Support with the charge date, amount, currency, and the last four digits of the payment method if needed. Never send a full card number, bank login, password, or security code.
A billing investigation does not guarantee a refund. Refund requests are reviewed under Severon's refund policy and must be submitted within the stated eligibility window.
---
Keeping your Severon account and API keys secure
Treat your Severon account, sign-in links, and API keys as private credentials.
Recommended security practices
- Never place API keys in public HTML, browser-side JavaScript, screenshots, or public repositories.
- Store server credentials in a protected secret manager or server-side environment.
- Do not share sign-in links, passwords, or verification codes.
- Use separate API keys for separate applications when available.
- Rotate a key immediately if it may have been exposed.
- Remove keys that are no longer used.
- Give team members only the access they need.
If a key is exposed, revoke or rotate it before contacting Support. Deleting it from a public repository may not be enough because copies can remain in history or caches.
Severon Support will never need your full password, payment-card number, or active API secret.
---
Troubleshooting slow responses or unavailable models
Response time can vary with the selected model, reasoning level, task complexity, tool use, and current demand.
If Severon is unusually slow:
1. Check the Severon status page for an incident.
2. Confirm that your internet connection is stable.
3. Refresh the page and retry once.
4. Start a new conversation if the current one contains a large amount of context.
5. Remove unnecessary files or instructions.
6. Select another suitable available model.
Do not repeatedly submit the same prompt while a response is still running, because this can create duplicate tasks and additional usage.
If the issue continues, contact Support with the model, approximate start time, browser, and any displayed error or request ID.
---
Reporting a problem to Severon Support
A clear report helps Support investigate quickly.
Include:
- What you were trying to do
- What you expected to happen
- What actually happened
- The exact error message
- The date, time, and time zone
- Your browser, device, and operating system
- The model or feature involved
- A request ID, if one was shown
- A screenshot with sensitive information removed
Before contacting Support, refresh the page, check the status page, and try the action once more if it is safe to do so.
Do not send passwords, sign-in links, verification codes, full payment-card numbers, private API keys, or confidential content. Support may ask for additional diagnostic information, but you should always remove secrets first.