News & Updates, AI Assistants, Industry Analysis

xAI Quietly Restricts Free Grok Access as Multi-Day Outage Persists

Free Grok users have been silently locked out of Expert and Auto models since April 24, following days of unacknowledged High Demand errors that left both free and paid users unable to access the chatbot.

3 min read
xAI Quietly Restricts Free Grok Access as Multi-Day Outage Persists

Image by xAI

xAI Quietly Restricts Free Grok Access as Multi-Day Outage Persists

Developers and users relying on Grok hit a wall this week. What started as sporadic "High Demand" errors on April 21 has stretched into a multi-day reliability crisis — and as of April 24–25, xAI quietly paywalled its best models without any public announcement.

Free users attempting to use Expert or Auto mode on grok.com are now redirected to a subscription page requiring a $30-per-month SuperGrok plan. The restriction appeared without notice in the model picker, and xAI has issued no statement about it. Paid subscribers report the change happened silently — there was no email, no status post, no X thread from the company.

StatusGator logged more than 400 user-submitted outage reports in a 24-hour window spanning the United States, Europe, and Asia. Third-party monitoring confirmed elevated reports throughout the week. The irony: xAI's official status page at status.x.ai listed services as "fully operational" throughout.

What Happened

The cascade of issues began around April 21, when xAI rolled out new Grok features including custom templates and expanded image capabilities — a feature drop that predictably drove a surge in usage. The same day, SpaceX announced a landmark deal with AI coding startup Cursor, securing the right to acquire it for $60 billion later this year or pay $10 billion for joint model training using xAI's Colossus supercomputer. The implication: significant GPU capacity on Colossus may have been redirected from inference to training.

Users began reporting the familiar "Grok is under high demand. Please try again later or upgrade your plan to get priority access" message, initially assuming it would clear within hours. It did not. Free users hit hard limits after as few as five to ten messages. Even paid SuperGrok subscribers encountered the same errors, raising questions about whether xAI was deliberately throttling lower tiers to push upgrades.

"How TF has Grok been unusable for 3–5 days, and status says fully operational, and xAI has not even acknowledged the complaints at this point??" one user posted on X, drawing hundreds of replies from developers in the same situation.

The Silent Paywall

The free-tier lockdown appears to have gone into effect on or around April 24–25, without any public announcement. Model options for unauthenticated users or non-paying accounts are now limited to the Fast variant — a lightweight option that lacks the reasoning depth of Expert mode, which had been available to free users. Clicking Expert or Auto redirects directly to the SuperGrok subscription page.

Community speculation on X and Reddit points to two overlapping explanations. First, Elon Musk posted on April 16 that Grok Build — xAI's dedicated coding terminal aimed at developers — would enter beta "next week." As of April 26, Grok Build has not launched. Second, with Grok 4.3 beta now running in parallel on SuperGrok Heavy accounts at $300/month, operating multiple model versions simultaneously places additional strain on shared infrastructure.

"Model is down for free users, is xAI saving compute for Grok Build," one developer account posted on X, attaching a screenshot of the grayed-out model options. xAI has not confirmed or denied this theory.

Developer API vs. Consumer Surface

Developers using the xAI API for backend integrations appear to have encountered a different timeline than consumer users on grok.com. Backend API services have generally recovered faster during prior demand spikes, though the current situation remains fluid. xAI reduced agent tool API pricing by up to 50% in April 2026, capping costs at $5 per 1,000 successful calls — a compelling number for agentic coding workflows that makes Grok an attractive inference option when it is actually available.

The Grok Voice Agent API went generally available earlier this month, claiming sub-1-second time-to-first-audio and first place on the Big Bench Audio benchmark. Whether the Voice API is also affected by the current capacity strain is not publicly confirmed.

For teams evaluating Grok as part of a production stack, the pattern this week — multi-day errors with no official acknowledgment, a silent tier restriction, and a status page showing green while hundreds of users report being locked out — is a meaningful reliability signal.

What's Unconfirmed

xAI has not issued any public statement about the free-tier model restriction or the extended "High Demand" errors. The company has not confirmed whether compute was redirected to SpaceX-Cursor model training or Grok 5's active training run on Colossus 2. There is no official launch date for Grok Build's public beta. xAI has not clarified whether the Expert and Auto model paywall is a permanent policy change or a temporary capacity management measure tied to current infrastructure pressure.

Share:

Other Latest News

Windsurf Becomes Devin Desktop With New Agent Command Center
AI Agents, News & Updates, Code Editors

Windsurf Becomes Devin Desktop With New Agent Command Center

Cognition has relaunched Windsurf as Devin Desktop, a full IDE with an Agent Command Center and open ACP protocol support that lets Codex, Claude Agent, OpenCode, and custom in-house agents run side by side in a shared Kanban workspace.

Jun 3, 2026
OpenAI Codex Lands on Amazon Bedrock for 5 Million Weekly Users
AI Agents, News & Updates, Code Editors

OpenAI Codex Lands on Amazon Bedrock for 5 Million Weekly Users

OpenAI's AI coding agent Codex is now generally available on Amazon Bedrock, giving developers at AWS-native organizations direct access to Codex through existing cloud infrastructure, procurement, and compliance workflows.

Jun 3, 2026
Anthropic Ships Claude Opus 4.8 With Dynamic Workflows and 3x Cheaper Fast Mode
AI Agents, News & Updates, Code Editors

Anthropic Ships Claude Opus 4.8 With Dynamic Workflows and 3x Cheaper Fast Mode

Anthropic releases Claude Opus 4.8 with improved agentic coding benchmarks, a 3x cheaper Fast Mode, and Dynamic Workflows that run hundreds of parallel subagents inside Claude Code.

May 29, 2026
MiniMax Revenue Doubles as M3 Promises 15x Long-Context Speed Boost
News & Updates, AI Assistants, Industry Analysis

MiniMax Revenue Doubles as M3 Promises 15x Long-Context Speed Boost

MiniMax annualized revenue doubled in two months, crossing 1M enterprise users. The company just teased M3's sparse attention architecture claiming 15.6x faster decoding at 1M-token contexts.

May 28, 2026
Anthropic Ships Free Claude Code Security Plugin to Catch Vulnerabilities in Real Time
AI Agents, News & Updates, Code Editors

Anthropic Ships Free Claude Code Security Plugin to Catch Vulnerabilities in Real Time

Anthropic's new Security Guidance plugin for Claude Code automatically scans code changes for injection flaws, unsafe deserialization, and 25+ dangerous patterns before they reach pull requests — free for all users on every plan.

May 27, 2026
Anthropic's Mythos Model Surfaces in Claude Code Ahead of Wider Release
AI Agents, News & Updates, Code Editors

Anthropic's Mythos Model Surfaces in Claude Code Ahead of Wider Release

Anthropic's most powerful restricted model, Claude Mythos 1, briefly appeared in Claude Code and Claude Security interfaces on May 25 — signaling the first commercial rollout of the model that found over 10,000 zero-day vulnerabilities under Project Glasswing.

May 26, 2026
← Scroll for more →