Model intelligence.
Follow the changes. Understand what comes next.
Why AI Router ZDR Labels Are Not a Guarantee: Inside OpenRouter's MiniMax Warning and a 4-Step Hardening Guide
Following Dragos's November 19, 2026 study of two million canary-seeded prompts, OpenRouter emailed affected customers on November 20–21 warning that prompts and outputs sent to MiniMax models from July to October 2026 may have been read by a third party—whether served directly or forwarded via AtlasCloud, Novita, or StreamLake—and removed ZDR labels from affected endpoints pending investigation.
Latest stories126
Newest publishedCodex Says "Selected Model Is at Capacity"? Here's What Users Report and What to Try
Since October 18, more than a dozen issues in OpenAI's Codex repository on GitHub have reported the error "Selected model is at capacity. Please try a different model." This article covers what those reports describe and what has worked for the users who filed the reports. OpenAI has not explained the error, and the OpenAI status page shows no incident.
What Ultracode Does in Claude Code, and Why It Uses More of Your Usage
Claude Code's documentation describes ultracode as a setting rather than a model effort level. With it on, Claude plans a dynamic workflow, a script that runs many subagents, for each substantive task. The documentation says this makes each request use more tokens, so subscription users reach their session or weekly limits sooner.
Why Claude Code Switches Models When a Message Is Flagged. Here's What You Can Do
Several Claude models in Claude Code run with safety classifiers. When one flags a request, Claude Code re-runs it on a fallback model, and the session stays there. Anthropic's docs describe a setting called switchModelsOnFlag that makes Claude Code ask first. Some users report in GitHub issues that the switch still happens with it turned off.
Where to Use DeepSeek V4.1 Flash for Free. WorkBuddy and Huawei's CodeArts Both Offer It
WorkBuddy Intl's free DeepSeek V4.1 Flash window ended October 31. Huawei Cloud CodeArts ended its limited-time free-token claim at 22:00 on October 20 (original claim pages now 404). DeepSeek Search and Fetch tool calls bill per request after 00:00 Beijing time on November 1, at $0.75 per 1,000 successful requests each; docs may still show an unsynced free-window line.
OpenAI Is Taking Applications for Its Codex Ambassador Program Until November 1. You Have to Be Picked to Get the Pro Subscription
OpenAI is accepting applications for its Codex Ambassador program until November 1. Ambassadors get a Pro subscription plus Codex and API credits during the six-month program, but they have to apply, be selected and contribute an event or piece of content every month. Students can't apply.
Gemini 2.5 Models Retire on Vertex AI on October 20, and Here's What Google Says to Use Instead
Google Cloud retires gemini-2.5-pro, gemini-2.5-flash and gemini-2.5-flash-lite on Vertex AI on October 20, 2026, and the replacements it lists for 2.5 Pro are both Flash models. On the Gemini API, the same three models have no shutdown date.
Custom GPTs and Claude Design Both Shut Down in December, and Not Everything Moves Over
Custom GPTs retire December 11 and standalone Claude Design closes December 14. Here's what moves over, what gets deleted and what to save first.
How to Connect Claude Code to Unreal Engine With Epic's Official Plugin
Epic's Claude Code plugin for Unreal Engine isn't new; it's been out since June. Here's how to install it, turn on Unreal MCP in the editor, and check that Claude Code can reach it, based on Epic's own docs.
Google Is Offering Up to 10,000 Researchers Six Months of Antigravity's Top Tier for Free
Google is taking applications from researchers at universities, nonprofit institutes and national labs for six months of Antigravity's highest usage tier and its Gemini 4 Argon model, which it says is free of charge. Up to 10,000 people will be selected.
Codex on Your Mac Might Be Quietly Filling Your Disk With Crash Dumps
Mac users report the Codex desktop app writing Crashpad crash dumps nonstop, in some cases hundreds of gigabytes. Here's what the GitHub thread says about the cause and affected macOS versions, how to check the folder, and what's safe to delete.
Anthropic Still Describes Claude Code Limits in Multiples, So Here's What Each Plan Gets
Anthropic meters Claude Code with a five-hour session limit and a weekly limit that are shared with the Claude apps, and it publishes multiples rather than numbers. Here is what each plan gets, what happened to the 50% weekly boost, how resets and usage credits work, what the limit messages mean, and a community-confirmed note that Agents SDK usage can drain the five-hour limit faster than Claude Code itself.
How to Install DeepSeek Harness and Use It With deepseek-flash
DeepSeek Harness is DeepSeek's open-source agent harness, available as a desktop app for macOS and Windows or as a Web UI launched from the command line. Here is how DeepSeek's docs say to install it, connect it to deepseek-flash, and why DeepSeek wants users on v0.2.1-rc.1 or later.