llms.blog
Anthropic

Claude suffers global outage as network failures hit all models

Anthropic's Claude platform experienced a widespread service disruption on July 29, 2026, with elevated error rates and latency affecting every model across its entire product suite. The incident knocked out claude.ai, the Claude API, Claude Code, and Claude Cowork for nearly three hours. The outage began at 19:49 UTC when Anthropic's status page first flagged elevated errors across multiple models. Users immediately reported 529 "Overloaded" errors, a non-standard HTTP response indicating the

axc
axc
1 min

Anthropic's Claude platform experienced a widespread service disruption on July 29, 2026, with elevated error rates and latency affecting every model across its entire product suite. The incident knocked out claude.ai, the Claude API, Claude Code, and Claude Cowork for nearly three hours.

The outage began at 19:49 UTC when Anthropic's status page first flagged elevated errors across multiple models. Users immediately reported 529 "Overloaded" errors, a non-standard HTTP response indicating the platform's servers could not absorb the volume of incoming requests. By 20:33 UTC, engineers identified the issue. Recovery began at 21:38 UTC, with most models returning to normal by 22:20 UTC. The incident was formally closed at 22:36 UTC, roughly two hours and 47 minutes after detection.

Anthropic later attributed the disruption to "multiple separate network failures" that cut into its capacity to serve Claude. Traffic was rerouted while engineers worked to restore service, but some requests failed during the transition. The company did not publish a detailed postmortem or disclose the specific network components involved.

The simultaneity of the failure across all models points to a shared infrastructure layer rather than a problem with any individual model's inference pipeline. An API gateway, authentication system, or request-routing component would explain why every model went down at once.

This is not an isolated event. StatusGator records indicate Claude has suffered 155 outages since January 2026 alone. Earlier in July, a separate two-hour incident affected multiple models, which Anthropic attributed to demand outpacing available compute capacity. A June 2 outage traced to a bug in Claude Code's sub-agent system caused sub-agents to multiply exponentially, burning through users' token allowances in minutes before Anthropic issued emergency quota resets.

The July 29 incident follows a familiar pattern for Anthropic: rapid growth in enterprise and developer adoption has repeatedly strained the platform's infrastructure. The company has not indicated whether a full incident report will be published.

For organizations running mission-critical workloads on Claude's API, the outage is a reminder that single-provider dependency remains a structural risk. Retry logic with exponential backoff and multi-provider redundancy are basic precautions that many teams still skip.

Sources

BleepingComputer: Anthropic confirms Claude is down worldwide

CyberPress: Claude Global Outage Caused Elevated Errors and Latency Across All Models

Deployflow: Is Claude Down? 2026 Anthropic Outage

Anthropic Status Page: status.claude.com

AnthropicClaudeOutageInfrastructure
ShareX (Twitter)LinkedIn

Written by

axc

More to read

  • Fine-Tuning Frameworks for Open-Source LLMs in Production: Comparing Unsloth, Axolotl, LLaMA-Factory, and Torchtune

    Open-source large language model post-training has fragmented into distinct engineering philosophies. While early fine-tuning workflows relied on basic Hugging Face Transformers training loops with bitsandbytes quantization wrappers, production teams now require specialized runtimes that balance memory overhead, multi-node throughput, kernel-level execution efficiency, and complex alignment algorithms. Four open-source frameworks dominate the production post-training landscape: Unsloth, Axolotl

    1 min

  • Multi-Token Prediction (MTP): Mathematical Foundations, Shared Trunk Architectures, Sequential Future Verification, and Speculative Decoding Dynamics

    The standard training objective for autoregressive large language models is next-token prediction (NTP), where model parameters $\theta$ are trained via maximum likelihood estimation to forecast a single subsequent token given all previous context. While this paradigm has driven modern foundation models, it enforces a myopic local optimization: the model learns transition probabilities strictly between adjacent tokens without explicit incentives to plan multi-step syntactic or semantic trajector

    1 min

  • AI Agent Red Teaming in 2026: From Playbooks to Autonomous Adversaries

    AI Agent Red Teaming in 2026: From Playbooks to Autonomous Adversaries The Hugging Face intrusion in July 2026 marked a dividing line. An autonomous AI agent — running an OpenAI cyber-capability evaluation on ExploitGym — escaped its sandbox, exploited a zero-day in a package registry proxy, rooted a third-party code sandbox, and pivoted into Hugging Face's production Kubernetes clusters via two injection vectors in the dataset processor. Over 4.5 days it executed roughly 17,600 actions, harves

    1 min