Don't stop here
Hand-picked guides our readers explore right after this one.
Expert guide to Claude prompts with XML tags, artifacts, and complex reasoning
Read the guideStunning image generation with Midjourney prompt mastery
Read the guideAI prompts for code generation, debugging, architecture, DevOps, and career growth
Read the guideIs Claude down right now, or is it just you? First, check status.anthropic.com and Downdetector — if Anthropic is having an outage or degraded performance, nothing on your end will fix it and you just have to wait. If the status page is green, the problem is usually local (your account, browser, or network) and clears in a few minutes. This guide covers both: how to confirm an outage today, what each Claude error means — including the overloaded error (529), API error 500, Claude Code's 'API Error: 400', and 'error sending message' — and the fixes that actually work across the web app, mobile, and Claude Code. (Claude Code issues usually track the same Anthropic status, so check it first.)
Claude down and need to keep building?
If you were using Claude to build a website or app, you don't have to wait out the outage. Lovable is a separate AI app builder that turns a prompt into a working, deployed full-stack app — so you can keep shipping. It's free to start.
Affiliate link, we may earn a commission at no extra cost to you.
claude.ai shows a blank page or infinite spinner on load
Messages send but no response appears, cursor just blinks
'Error sending message' or 'Something went wrong. Please try again.'
'Overloaded' / 'Claude is experiencing high demand right now' (API error 529)
API error 500 ('Internal server error') from the API or an app built on Claude
Claude Code shows 'API Error: 400' or 'API Error: 529' in the terminal
Claude Code feels slow or gives worse answers — degraded performance today
'This model isn't available right now. You can switch to another model to continue using Claude.'
'Claude couldn't finish this response. Try again in a moment.'
'We couldn't connect to Claude. Please check your network connection and try again.'
'A previous response is still running in this conversation. It will appear shortly — try again once it finishes.'
'Unfortunately, Claude is not available to new users right now. We're working hard to expand our availability soon.'
Claude stops mid-response and never finishes generating
Claude Code hangs, errors, or won't connect in your terminal/IDE
Claude mobile app (iOS/Android) crashes on launch or mid-chat
Anthropic experiences periodic degraded service, especially during high-demand periods or right after major model releases (Opus, Sonnet, and Haiku launches have caused widespread slowdowns). When capacity is maxed out, the API returns error 529 ('overloaded') and claude.ai shows a high-demand banner — it's Anthropic-side and clears on its own. Check status.anthropic.com before any local troubleshooting.
A 500 ('api_error') is an internal server error inside Anthropic's infrastructure, separate from overload. It's transient and usually resolves on a retry; if it persists for everyone, it's an incident and will appear on status.anthropic.com. You can't fix a 500 locally — retry with a short backoff.
Claude's web app stores session state aggressively. After Anthropic deploys updates (common on weekdays), stale cache data causes blank screens, infinite loading, broken UI, or 'error sending message'. This is the most common cause of 'it was working yesterday' reports.
Content-modifying extensions, ad blockers (uBlock Origin, AdGuard), privacy tools (Privacy Badger, Ghostery), and reading-mode extensions frequently break claude.ai's interface. Claude's frontend is more sensitive to DOM manipulation than ChatGPT's.
Claude's paid plans use a rolling session limit that resets roughly every 5 hours, plus weekly caps introduced in 2025 (Pro and Max plans each have their own ceilings; the free tier is stricter). When you hit either limit, Claude shows a banner and refuses new messages until the window resets — claude.ai displays a countdown so you know when access returns. Hitting the weekly cap is a common reason heavy users find Claude 'not working' mid-week even though the status page is green.
In Claude Code, 'API Error: 400' is an invalid_request_error — the request the CLI sent was rejected, often because the conversation grew too large, a message got into a bad state, or the CLI/IDE extension is out of date. It is different from a 529 (overload) or 500 (server error): a 400 is about the request itself, so /compact, a fresh session, or updating Claude Code usually clears it. Claude Code runs on Anthropic's API, so a wide outage there also shows up as 529s and connection errors in your terminal.
Corporate networks, school networks, and some VPN configurations block claude.ai or Anthropic's API endpoints. Common in enterprise environments that whitelist AI tools individually. Symptoms: requests hang indefinitely rather than returning a clean error.
The Claude iOS/Android apps and the Claude Code CLI update frequently. Running an old version after Anthropic makes backend changes causes crashes, blank responses, sync failures, and avoidable 400s. App stores and package managers don't always auto-update immediately.
When to try: Always first, before any local troubleshooting
Go to status.anthropic.com. Look for any active incidents or degraded performance notices (it covers Claude.ai, the API, and Claude Code separately). Also search X/Twitter for 'Claude down' to get real-time user reports — they precede official status updates by 10-20 minutes. If there's an active incident, wait for Anthropic to resolve it; no local fix will help.
When to try: As soon as you see a specific error code or message
Read the exact error before acting — it tells you whose problem it is. 529 ('overloaded') means Anthropic is at capacity: wait and retry with a short backoff, or switch to a lighter model (Haiku) which is often less loaded. 500 ('api_error') is a transient server error on Anthropic's side: retry; if it persists for everyone, check the status page. 400 ('invalid_request_error', common in Claude Code) means the request itself was rejected — that's local and fixable (see the Claude Code step). 'Error sending message' on claude.ai is usually a transient network/server hiccup — just resend.
When to try: First local fix, 30 seconds, no data loss
Press Ctrl+Shift+R (Windows/Linux) or Cmd+Shift+R (Mac) in your browser to bypass the cache and force a full reload of claude.ai. This resolves most 'blank screen after update' and 'error sending message' issues without needing to clear all cookies. Try this before the more disruptive cache-clear step.
When to try: If hard refresh didn't resolve the issue
In Chrome: Settings → Privacy and security → Third-party cookies → See all site data and permissions → search for 'claude' and 'anthropic' → Delete all. In Firefox: Address bar, click the lock icon → Clear cookies and site data. In Safari: Settings → Privacy → Manage Website Data → search 'anthropic' → Remove. Sign back in after clearing.
When to try: To isolate browser extension conflicts
Open an incognito window and go to claude.ai. If it works there but not your normal browser, a browser extension is causing the issue. Return to your normal browser and disable extensions one by one, starting with ad blockers and privacy tools, as these most commonly break Claude's interface.
When to try: If Claude loads fine but refuses to respond to new messages
Sign in to claude.ai and look for a usage banner near the message box. If you've hit a limit, Claude shows a timer for when your next window opens — the session limit resets roughly every 5 hours, but there's also a weekly cap that, once hit, blocks you for the rest of the week even though everything else works. Free users have stricter limits; Pro and Max raise both ceilings. If Claude loads fine but refuses new messages and the status page is green, this is almost always why.
When to try: For Claude Code errors in the terminal or IDE
A 400 in Claude Code is a rejected request, not an outage. Run /compact to shrink an over-long conversation, or start a fresh session to clear bad state. Update the CLI (npm i -g @anthropic-ai/claude-code@latest or your installed method) since old versions throw avoidable 400s. If you instead see 529s, connection failures, or 'degraded performance', that's the API — check the API component on status.anthropic.com, then verify your network/proxy and that your credits or plan are active. The Anthropic GitHub issues for Claude Code are a good place to confirm whether others are hitting the same error today.
When to try: For Claude mobile app issues specifically
For iOS: App Store → your account icon → scroll to Claude → Update. For Android: Play Store → My apps → Claude → Update. If already up to date and still broken, delete the app and reinstall it fresh. This clears all cached data and corrupted local state. You won't lose conversation history since it's stored server-side.
When to try: If the app loads but requests always hang or time out
Switch from WiFi to mobile data (or vice versa) to test whether your network is the issue. Disable VPN if active. If on a corporate or school network, try from a personal device on mobile data. Claude is sometimes blocked by network-level content filters before the error reaches your browser.
When to try: If all other fixes fail and you need to keep working
Try claude.ai in Chrome if you normally use Safari or Firefox (or vice versa). Claude works best on modern Chromium browsers. If it works on a different device, the problem is local to your original browser profile. Create a new browser profile with no extensions to confirm.
Bookmark status.anthropic.com, it's the fastest way to confirm outages
Keep the Claude mobile app updated to the latest version to avoid compatibility issues
Use a separate browser profile for Claude if you rely heavily on extensions in your main profile
If you use Claude for critical work, track both your ~5-hour session window and your weekly cap so you don't get blocked mid-project; keep a fallback model handy for overload (529) periods
Claude Projects sync conversation history server-side, you won't lose context if you need to clear local cache
Contact Anthropic support via support.anthropic.com if: (1) Your account shows an unexpected suspension, (2) Billing issues persist after verifying your payment method, (3) You've tried all fixes and the issue affects your account specifically while status.anthropic.com shows all systems operational, (4) You believe your Pro limits are being applied incorrectly. For general service issues, self-service troubleshooting resolves the vast majority of cases.
Incomplete responses happen for a few reasons: (1) Network interruption during streaming, (2) Response hit Claude's output length limit (very long generations sometimes cut off), (3) Content policy filter triggered partway through, (4) Server-side timeout on Anthropic's end during high load. Fix: refresh the page and rephrase the request into shorter steps. If asking for something very long, break it into parts.
Check status.anthropic.com for official status. For real-time user reports (which often precede official updates), search 'Claude down' or 'claude.ai not working' on X/Twitter. Downdetector.com also tracks user-reported Claude issues. Anthropic typically acknowledges major incidents within 15-30 minutes of onset.
Claude's paid plans use a rolling session limit that resets roughly every 5 hours, plus a weekly cap added in 2025. The exact number of messages adapts to demand and to which model you use — heavier models like Opus 4.8 (and the most capable model, Fable 5) consume your allowance faster than Sonnet 4.6 or Haiku 4.5. Free users have stricter limits; Pro and Max raise both ceilings. The claude.ai interface shows a countdown to your next window. If you're blocked mid-week with the status page green, you've likely hit the weekly cap rather than the 5-hour one.
Yes. Claude has native apps for iOS (App Store) and Android (Play Store) as of 2024. The mobile apps support full conversation history, Projects, and file uploads. Common mobile issues: app crashes after updates (fix: reinstall), slow responses on cellular (fix: switch to WiFi), and login loops (fix: sign out fully and back in).
The web interface (claude.ai) and API use the same underlying models but have slightly different system prompts and context configurations. Web Claude has a 'helpful assistant' system prompt built in; raw API calls start without one. Output format also differs since web is optimized for markdown rendering. This is expected behavior, not a bug.
Claude's error messages are generally more transparent, it tells you directly when you've hit a usage limit and shows you the reset timer. ChatGPT tends toward generic 'Something went wrong' messages. Claude's status page (status.anthropic.com) is also usually updated faster. However, Claude's usage limits (8-hour windows) differ from ChatGPT's model-specific rate limits.
No. Claude is a cloud-only service with no offline mode. All inference runs on Anthropic's servers. There's no local version of Claude available to end users (unlike some open-source models you can run locally). You need an active internet connection for every message.
Claude speed varies with server load. Peak times (US weekday afternoons, and the hours right after a major model release) cause noticeably slower responses, and at the extreme you'll see overload errors (529). Model choice also matters: Haiku 4.5 is the fastest, Sonnet 4.6 balances speed and capability, and Opus 4.8 — along with Fable 5, the most capable model — reasons longer and so feels slower on hard tasks. If speed matters right now, switch to a lighter model for simpler work. Verified June 2026.
Check two sources to tell the difference. If status.anthropic.com shows an active incident or Downdetector shows a spike in 'Claude down' reports, it's an outage on Anthropic's side and you just have to wait. If both are clear but Claude still fails only for you, it's local — hard-refresh (Ctrl/Cmd+Shift+R), clear claude.ai site data, try incognito to rule out extensions, or switch network. Outages affect everyone; local issues affect only you.
The official source is status.anthropic.com, which covers Claude.ai, the API, and Claude Code, with incident history and component-level status. For faster, real-time signal, Downdetector and searching 'Claude down' on X/Twitter often surface an outage 10–20 minutes before the status page updates. Bookmark status.anthropic.com so you can confirm at a glance.
Claude Code runs on Anthropic's API, so when Claude Code can't connect or errors out, it usually reflects the same Anthropic incident shown on status.anthropic.com (check the API component specifically). If the status page is green but Claude Code still fails, check your API key/credits, your network or proxy, and that your CLI/IDE extension is updated. A wide Anthropic outage takes Claude.ai, the API, and Claude Code down together.
It means the specific Claude model you selected is temporarily unavailable — usually due to high demand or a partial incident affecting that model. The message itself suggests the fix: switch to another model (e.g. from Opus to Sonnet) to keep working, then switch back later. If no model is available, check status.anthropic.com for a broader incident.
These are transient errors. 'Claude couldn't finish this response — try again in a moment' usually means a server-side timeout or interruption during generation under load; just resend, ideally in smaller steps. 'We couldn't connect to Claude — check your network' points to a connection problem on your side (network, VPN, firewall) or a brief outage; switch network, disable VPN, and retry. If they persist for everyone, check the status page.
The message 'Unfortunately, Claude is not available to new users right now. We're working hard to expand our availability soon' appears during periods when Anthropic temporarily limits new sign-ups due to capacity. It's not an error you can fix — existing users are usually unaffected. Try again later, or sign in if you already have an account. It typically eases as Anthropic scales capacity.
Error 529 is Anthropic's 'overloaded_error' — Claude's servers are temporarily at capacity, usually during peak hours or right after a new model launch. It is not a problem with your account, request, or network, and there's nothing to fix on your end. Wait a moment and retry; retrying with a short, increasing backoff works best. If you're on the API or in Claude Code, switching to a lighter model like Haiku 4.5 (which is often less loaded) can get you unblocked. A wave of 529s for everyone usually shows up on status.anthropic.com.
A 500 is 'api_error' — an internal server error inside Anthropic's infrastructure, distinct from a 529 overload. It's typically transient: retry the request, ideally with a short backoff, and it usually succeeds. If 500s persist across requests and accounts, it's an incident — check status.anthropic.com. Because it originates server-side, you can't fix a 500 by changing your prompt, key, or network.
In Claude Code, 'API Error: 400' is an invalid_request_error — the request the CLI sent was rejected, not an outage. The usual causes are a conversation that grew too large, a message stuck in a bad state, or an out-of-date CLI. Fixes, in order: run /compact to shrink the context, start a fresh session to clear bad state, and update Claude Code to the latest version (e.g. npm i -g @anthropic-ai/claude-code@latest). If you see 529 or connection errors instead of 400, that's the API being overloaded or down — check the API status rather than your session.
'Error sending message' on claude.ai is almost always a transient hiccup — a brief network drop, a momentary server error (500), or overload (529). First, just resend the message; it often goes through on the second try. If it keeps failing: hard-refresh (Ctrl/Cmd+Shift+R), check your internet connection, disable VPN, and try an incognito window to rule out a browser extension. If it's failing for everyone, check status.anthropic.com for an incident.
Claude doesn't 'crash' the way a desktop app does — when it's unavailable it's almost always either an Anthropic-side incident or overload (error 529), or a local issue with your browser/account. To tell which: if status.anthropic.com shows an active incident or Downdetector shows a spike in reports, it's down for everyone and you wait it out. If both are clear, it's local — hard-refresh, clear claude.ai site data, try incognito, or switch network. Searching 'Claude down' on X/Twitter is the fastest way to confirm a today outage in real time.
Claude Code runs on Anthropic's API, so 'degraded performance' there usually mirrors an API-side incident — check the API component on status.anthropic.com, which posts degraded-performance notices separately from full outages. During these windows you may see slower responses, more 529 overload errors, or retries. It's temporary and not something you fix locally; if you need to keep moving, lower the effort/model for routine work. If the status page is green but only your sessions are slow, update Claude Code and check your network/proxy.
Anthropic maintains the Claude Code repository and issue tracker on GitHub, and it's a useful place to confirm whether others are hitting the same error today and whether a fix is in progress. Search the issues for your exact error string (e.g. 'API Error: 400' or '529'). For service status it's still faster to check status.anthropic.com, but GitHub issues are the best signal for CLI-specific bugs, regressions after an update, and workarounds posted by the maintainers and community.
If you were using Claude to build a website or app and you're blocked by an outage, you don't have to stall. Lovable is a separate AI app builder that turns a plain-English prompt into a deployed full-stack app (with a database, auth, and a live URL) — a practical way to keep shipping while you wait. It has a free plan, so you can pick up where you left off. Note that during a very wide Anthropic outage some AI tools can be affected generally; Lovable is its own platform and is often available, and it's a strong alternative builder regardless.