Claude Code's $200 Monthly Price Sparks Revolt as Block's Free Goose Gains Traction
SAN FRANCISCO — Anthropic's Claude Code, a terminal-based AI agent capable of autonomously writing, debugging and deploying code, has drawn widespread developer interest but also sharp criticism over its subscription costs reaching up to $200 per month. In response, Block's open-source Goose project has surged in popularity by delivering similar functionality at no cost while running locally on users' machines.
The contrast highlights growing tension in the AI coding tools market between premium cloud services and free, privacy-focused open-source alternatives. Goose, developed by the financial technology company formerly known as Square, now exceeds 26,100 GitHub stars with 362 contributors and 102 releases. Its latest version, 1.20.1, was released Jan. 19, 2026.
Anthropic's Rate Limits Fuel Developer Frustration
Anthropic offers Claude Code through tiered subscriptions. The free plan provides no access. The Pro tier costs $20 monthly or $17 with annual billing, restricting users to 10-40 prompts every five hours — limits many developers report exhausting quickly during intensive sessions.
Higher Max plans at $100 and $200 per month provide more capacity: 50-200 prompts and 200-800 prompts respectively, along with access to Anthropic's Claude 4.5 Opus model. In late July, the company introduced weekly rate limits, allocating Pro users 40-80 hours of Sonnet 4 usage per week and $200-tier Max users 240-480 hours of Sonnet 4 plus 24-40 hours of Opus 4.
Developers have criticized these "hours" as opaque token-based limits that fluctuate based on codebase size and task complexity. Independent analysis estimates actual limits at roughly 44,000 tokens for Pro users and 220,000 tokens for the top Max plan. Complaints on Reddit and developer forums describe hitting daily caps within 30 minutes of heavy use, prompting some cancellations.
Anthropic has stated the restrictions impact fewer than 5% of users and target those running Claude Code continuously in the background. The company has not specified whether this percentage applies to all users or only Max subscribers.
Goose Offers Local, Model-Agnostic Alternative
Block's Goose takes a different approach as an "on-machine AI agent." Unlike Claude Code, which relies on Anthropic's cloud servers, Goose operates locally using open-source models downloaded by the user.
The tool goes "beyond code suggestions" to install, execute, edit and test code with any large language model. Users can connect Goose to cloud APIs from Anthropic, OpenAI's GPT-5, Google's Gemini, Groq or OpenRouter. For fully offline operation, it integrates with tools like Ollama to run models directly on local hardware.
This setup eliminates subscription fees, usage caps and rate limits while keeping code and data on the user's device. "Your data stays with you, period," said Parth Sareen, a software engineer who demonstrated Goose in a recent livestream. The ability to work offline — including on airplanes — has proven particularly appealing.
Impact on Developers and the AI Coding Market
For developers, Goose provides a genuine zero-cost option with comparable core capabilities to Claude Code. Those prioritizing the highest model quality and willing to accept restrictions may continue using Anthropic's service. Others seeking cost savings, privacy, offline access and flexibility now have a viable alternative, though it requires hardware capable of running local models and some setup overhead.
The project's rapid growth — from launch to more than 26,100 GitHub stars — reflects strong demand for autonomous coding agents that respect developer control. It also signals the maturing capabilities of open-source AI infrastructure, allowing sophisticated tools to compete with well-funded commercial offerings.
Block's involvement lends credibility to the effort. The company, led by Jack Dorsey, has invested in open-source initiatives alongside its core payments business.
What's Next
Goose's development pace, with 102 releases since launch, suggests continued rapid iteration. The project's model-agnostic design positions it to benefit from improvements in open-source LLMs, potentially narrowing any performance gap with premium cloud models.
For Anthropic, sustained developer backlash may pressure adjustments to pricing and rate limit transparency. The company has not announced changes as of the latest reports.
The emergence of Goose underscores a broader industry trend: as open-source AI tools advance, commercial providers face increasing competition from free alternatives that prioritize user autonomy and data privacy. Developers now have a clear choice between premium convenience and open-source flexibility when selecting AI coding agents.
(Word count: 748)