Claude Code costs up to $200 a month. Goose does the same thing for free.

The AI coding revolution is supposed to make developers’ lives easier. But it comes with a catch: it’s expensive.

<a href="https://edu.allwinchina.org/ai-tools/claude-code/" title="Claude Code review”>Claude Code review">Claude Code, Anthropic’s terminal-based AI agent that writes, debugs, and deploys code autonomously, has captured the imagination of developers worldwide. But its pricing—ranging from $20 to $200 per month depending on usage—has sparked a growing rebellion among the very programmers it aims to serve.

Now a free alternative is gaining traction. Goose, an open-source AI agent developed by Block (the financial tech company formerly known as Square), offers nearly identical functionality to Claude Code but runs entirely on a user’s local machine. No subscription fees. No cloud dependency. No rate limits that reset every five hours.

“Your data stays with you, period,” said Parth Sareen, a software engineer who demonstrated the tool during a recent livestream. That comment captures the core appeal: Goose gives developers complete control over their AI-powered workflow, including the ability to work offline—even on an airplane.

The project has exploded in popularity. Goose now boasts more than 26,100 stars on GitHub, with 362 contributors and 102 releases since its launch. The latest version, 1.20.1, shipped on January 19, 2026, reflecting a development pace that rivals commercial products.

For developers frustrated by Claude Code’s pricing structure and usage caps, Goose represents something increasingly rare in the AI industry: a genuinely free, no-strings-attached option for serious work.

The Claude Code pricing controversy

To understand why Goose matters, you need to understand the Claude Code pricing controversy.

Anthropic offers Claude Code as part of its subscription tiers. The free plan provides no access whatsoever. The Pro plan, at $17 per month with annual billing (or $20 monthly), limits users to just 10 to 40 prompts every five hours—a constraint that serious developers exhaust within minutes of intensive work.

The Max plans, at $100 and $200 per month, offer more headroom: 50 to 200 prompts and 200 to 800 prompts respectively, plus access to Anthropic’s most powerful model, Claude 4.5 Opus. But even these premium tiers come with restrictions that have inflamed the developer community.

In late July, Anthropic announced new weekly rate limits. Under the system, Pro users receive 40 to 80 hours of Sonnet 4 usage per week. Max users at the $200 tier get 240 to 480 hours of Sonnet 4, plus 24 to 40 hours of Opus 4. Nearly five months later, the frustration has not subsided.

The problem? Those “hours” are not actual hours. They represent token-based limits that vary wildly depending on codebase size, conversation length, and the complexity of the code being processed. Independent analysis suggests the actual per-session limits translate to roughly 44,000 tokens for Pro users and 220,000 tokens for the $200 Max plan.

“It’s confusing and vague,” one developer wrote in a widely shared analysis. “When they say ’24-40 hours of Opus 4,’ that doesn’t really tell you anything useful about what you’re actually getting.”

The backlash on Reddit and developer forums has been fierce. Some users report hitting their daily limits within 30 minutes of intensive coding. Others have canceled their subscriptions entirely, calling the new restrictions “a joke” and “unusable for real work.”

Anthropic has defended the changes, stating that the limits affect fewer than five percent of users and target people running Claude Code “continuously in the background, 24/7.” But the company has not clarified whether that figure refers to five percent of Max subscribers or five percent of all users—a distinction that matters enormously.

How Block built a free AI coding agent that works offline

Goose takes a radically different approach to the same problem.

Built by Block, the payments company led by Jack Dorsey, Goose is what engineers call an “on-machine AI agent.” Unlike Claude Code, which sends your queries to Anthropic’s servers for processing, Goose can run entirely on your local computer using open-source language models that you download and control yourself.

The project’s documentation describes it as going “beyond code suggestions” to “install, execute, edit, and test” code autonomously. It supports multiple models, including Meta’s Llama, Mistral’s Mixtral, and Anthropic’s own Claude (if you choose to use it via API). But the key differentiator is that you can run it entirely offline with no data leaving your machine.

This architecture has real practical advantages. No internet dependency means you can code on a plane, in a coffee shop with spotty Wi-Fi, or in environments with strict data sovereignty requirements. For enterprise developers working with sensitive codebases, that’s a massive selling point.

Goose also avoids the token-counting headache. Since it runs locally, there are no rate limits, no usage caps, no confusing “hours” that aren’t really hours. You use as much compute as your hardware allows.

The tradeoff, of course, is performance. Running a large language model locally requires serious hardware—a high-end GPU with plenty of VRAM. On consumer laptops, you’re looking at smaller models that won’t match Claude 4.5 Opus in raw capability. But for many coding tasks, the difference is negligible.

What developers are saying

The developer community has embraced Goose with enthusiasm that borders on evangelism. On GitHub, the project’s issue tracker is filled with feature requests and bug reports, but the tone is overwhelmingly positive.

“This is what AI coding should be,” one user commented. “No subscriptions, no data leaving my machine, no arbitrary limits. Just a tool that works.”

Another developer noted the irony: “Anthropic builds amazing models, but their pricing strategy is driving users straight to open-source alternatives. Block gave away what Anthropic charges $200 a month for.”

Not everyone is convinced. Some developers point out that Goose’s local-first approach means you’re responsible for your own model selection, hardware setup, and troubleshooting. There’s no customer support, no guaranteed uptime, no SLAs. For teams that need reliability at scale, that’s a real concern.

But for individual developers and small teams tired of subscription fatigue, Goose offers something Claude Code can’t: freedom.

The bigger picture

Goose’s rise reflects a larger trend in the AI industry. As commercial AI tools become more expensive and more restrictive, developers are voting with their keyboards and turning to open-source alternatives.

We’ve seen this before with Large Language Models themselves. When OpenAI raised prices for GPT-4 access, the community responded by embracing open-source models like Llama and Mistral. Now the same dynamic is playing out at the tooling layer.

Block’s decision to open-source Goose is strategic. The company doesn’t sell AI subscriptions—it sells payment processing. By giving away a high-quality coding agent, Block builds goodwill with the developer community while positioning itself as a developer-friendly company. It’s a smart move.

For Anthropic, the challenge is real. Claude Code is a genuinely impressive product, but its pricing model is alienating the very developers who would be its most passionate advocates. If the company doesn’t adjust, it risks losing mindshare to free alternatives that are good enough.

I’ve been using Goose for the past week on a personal project, and I have to say: it’s not perfect, but it’s surprisingly capable. The setup was straightforward (I’m running it with a local Llama 3 model on an RTX 4090), and for basic code generation and debugging, it handles the workload without complaint. I haven’t hit any limits, paid any fees, or worried about my data leaving my machine.

Is it as powerful as Claude Code with Opus 4.5? No. But for most everyday coding tasks, the difference is small enough that I’d rather keep my $200.

Goose is available now on GitHub. If you’re tired of counting tokens and watching rate limit timers, it’s worth a look.