OpenAI and Anthropic roll out holiday usage boosts for AI developers
What looks like festive generosity also exposes growing pressure around AI pricing and usage limits for serious users.
Here’s something you don’t see every day: two bitter rivals making the exact same move almost on the same day.
Before December 25th, Anthropic announced it would double usage limits for Claude Pro and Max subscribers through December 31st. Hours later, OpenAI followed with its own announcement—Codex users would get increased usage limits through January 1st, complete with a festive twist: GPT-5.2-Codex-XMas, a special holiday version of their coding model that “performs exactly the same as GPT-5.2-Codex, just with a needed personality upgrade for the season.”
Both companies framed it as a holiday thank-you. Both picked the same week. Both addressed the same pain point. That’s not a coincidence. That’s a signal. What looks like generosity is actually an admission: the current pricing model for AI tools isn’t working for the people who use them most. And the companies building these tools know it.
Damilare Odedina
What Changed and Who Got What
Anthropic’s offer is straightforward. All Claude Pro ($20/month) and Max subscribers ($100-$200/month) receive double their usual usage limits from December 25th through December 31st. The expansion applies across Claude’s web app, Claude Code, and Claude in Chrome, with all weekly caps reset at the start of the promotion.
OpenAI’s approach is more selective. President Greg Brockman announced increased Codex usage limits until January 1st, targeting the company’s AI coding assistant users. The boost—delivered via the new Santa Codex model—applies to Codex users across different subscription tiers, but excludes standard ChatGPT users entirely.
That exclusivity sparked immediate frustration. One user captured the sentiment perfectly, joking they felt like “a Victorian orphan watching from the window” while Codex users enjoyed expanded access.
The numbers tell the story. Claude Pro users can send roughly 200 messages every eight hours before hitting limits. Max users get more, but even they report hitting weekly caps on intensive projects. OpenAI’s Codex faces similar complaints, with users reporting rate limits after just a few hours of active development work.
Credit: Claude
The Response Revealed a Bigger Problem
For developers who got the boost, the reaction was immediate relief. They could finally work on fixing bugs and refactoring without constantly watching their usage meters.
This holiday temporary offers sparked a bigger conversation: why can’t this just be the default?
Multiple users across both platforms called for permanent “Double Pro” pricing tiers—something between standard Pro plans and expensive Max subscriptions. The ask makes economic sense. Claude Pro at $20/month feels restrictive for serious daily use. Max at $100-$200/month is too just too expensive for hobbyists and indie developers who need more than casual access but can’t justify enterprise spending.
The frustration is real. Back in August 2025, Anthropic upset a lot of power users by putting weekly limits on their $200 Max plan—people definitely felt shortchanged. Even though it only affected a small group, it sparked a real debate about what “unlimited” actually means for AI subscriptions.
Why This Matters Beyond Holiday Cheer
These temporary expansions aren’t just for goodwill gestures. They’re market research disguised as holiday gifts.
Both companies are trying to figure out a major balancing act: can they double those usage limits to keep users happy without completely crashing their systems? If the answer is yes—and usage patterns remain sustainable—expect permanent tier adjustments in 2025.
The holiday boost also reveals something about infrastructure capacity. If both companies can afford to double limits temporarily, it suggests their systems can handle significantly more load than current pricing structures allow. These companies just proved they could offer more—they’re just choosing not to, at least not permanently.
Emmanuel Umahi
The $20-$200 Gap Is Where the Pain Lives
What’s emerging is a fundamental tension in AI pricing that no one has solved yet.
AI companies face massive infrastructure costs—training and running models requires enormous computational resources. But users are increasingly frustrated by unpredictable limits and arbitrary caps that interrupt workflows exactly when they’re most engaged.
Serious hobbyists, indie builders, and small teams who actually use this tool get stuck in the $20-$200 gap. Standard Pro plans feel too restrictive for anyone doing serious work. Max plans feel too expensive for anyone who isn’t a professional developer or enterprise user.
If these holiday experiments succeed, 2025 will likely bring a permanent mid-tier plan—something offering 3-5x the capacity of standard Pro at $50-$75/month. That would capture the serious hobbyist and indie developer market without requiring them to jump to $200/month enterprise pricing.
As one Reddit user put it: “I don’t mind paying for a great product. I just want to know what I’m getting and not hit an arbitrary wall in the middle of my work.”
That sentiment—repeated across forums, social media, and support tickets—is what both OpenAI and Anthropic are responding to with these temporary boosts. They’re acknowledging the problem without committing to a permanent solution.
But temporary fixes won’t work forever. As AI tools become more deeply integrated into daily workflows, users will demand pricing that matches how they actually work: predictable, transparent, and scaled to their needs.
The companies that figure this out first will win. The ones that don’t will watch their most valuable users—the serious builders who actually push these tools to their limits—walk away to competitors who offer what they need.
For now, even Santa Codex can’t solve the pricing problem. But the Victorian orphans won’t stay outside forever.
Emmanuel Umahi
