TL;DR
OpenClaw creator Peter Steinberger spent more than $1.3 million on OpenAI API tokens in a single month, primarily using GPT-5.5 for AI-driven development automation. The expenditure underscores the high costs associated with large-scale AI automation projects.
Peter Steinberger, the Austrian developer behind the open-source OpenClaw project, posted a screenshot showing his team spent over $1.3 million on OpenAI API tokens in just 30 days. The expenditure, primarily on GPT-5.5, highlights the significant costs involved in AI-driven software automation at scale.
Steinberger’s OpenClaw project employs roughly 100 Codex instances operated by a team of three, automating tasks such as pull request reviews, security scans, issue deduplication, and feature development. The reported bill covers 603 billion tokens across 7.6 million requests, with the highest daily spend reaching nearly $20,000.
The API costs are driven predominantly by the use of GPT-5.5 in ‘Fast Mode,’ which significantly increases token consumption. Steinberger clarified that disabling Fast Mode would reduce the bill to approximately $300,000, still a substantial figure. The project’s open-source nature means the costs are borne by OpenAI, not Steinberger personally.
Why It Matters
This development underscores the potential financial scale of AI-assisted development when used intensively. It reveals the gap between typical developer costs and the compute resources required for large-scale automation, raising questions about the economics of AI in software engineering and the sustainability of such high expenditures.
Agentic AI Architectural Patterns: Engineering Blueprint to Build 24/7 Autonomous Agents That Work While You Sleep | Master Production-Grade Automation, Build Deterministic Pipelines & Control Costs
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Background
OpenClaw has been a high-profile project in AI-assisted development, known for its autonomous agents managing code repositories. The project has faced scrutiny over its public impact and costs, especially as competitors like Nvidia develop their own AI tools. The recent spending figures shed light on the actual compute costs involved in running such AI automation at scale.
“The $1.3 million figure reflects Codex’s ‘Fast Mode’ pricing, which consumes credits at a significantly higher rate than standard execution.”
— Peter Steinberger
“Everything we build remains open source. This spending is research into how software development would change if token costs weren’t a constraint.”
— Steinberger
large-scale API token management software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What Remains Unclear
It remains unclear how sustainable or typical such high API costs are for other projects, and whether OpenAI will alter pricing or usage policies in response. The long-term implications for AI-driven development economics are still uncertain.
The AI Advantage for Software Developers: Prompts, Agent Systems, and High-Performance Workflows to Grow Faster in the Age of AI
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What’s Next
Further analysis is expected on the actual costs of AI automation at scale, potential shifts in API pricing, and how other developers and companies might approach high-volume API usage. Monitoring OpenAI’s policy updates and industry responses will be key.
Claude AI Office Mastery: A Professional Textbook for Emails, Meetings, Documents, Research, Automation, and Daily Productivity
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
Why did Steinberger’s team spend so much on API tokens?
The team used GPT-5.5 in ‘Fast Mode,’ which consumes tokens at a much higher rate, to maximize automation and testing in their open-source project.
Is this kind of spending typical for AI development?
No, most developers spend significantly less; Steinberger’s usage is at the extreme end, driven by the project’s scale and automation needs.
What does this mean for the future of AI-assisted coding?
The high costs highlight the need for more efficient models or pricing structures if AI-assisted development is to become widely sustainable.
Could OpenAI change its pricing to reduce costs?
OpenAI could adjust pricing, but it has not announced any such plans; the current costs reflect high-volume, high-performance use cases.
