Changes to GitHub Copilot Individual plans

2026-04-22 · Source: Simon Willison's Weblog · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Fundamental Awareness, quick

Summary

GitHub has announced significant changes to its Copilot Individual plans, including tightening usage limits, pausing new sign-ups for individual plans, and restricting access to Claude Opus 4.7 to the more expensive $39/month "Pro+" plan. Previous Opus models have been dropped entirely. These adjustments are attributed to the increased compute demands of agentic workflows, which consume substantially more resources than the original plan structure was designed to support. Unlike its previous per-request billing, Copilot is now implementing token-based usage limits on a per-session and weekly basis to manage these costs. The changes affect GitHub Copilot CLI, cloud agent, code review features on GitHub.com, and IDE integrations for VS Code, Zed, and JetBrains.

Key takeaway

For engineering leaders evaluating AI coding assistants, understand that agentic workflows fundamentally alter compute economics. Your teams should anticipate increased costs and tighter usage limits from providers like GitHub Copilot, necessitating a re-evaluation of budget allocations and potentially exploring alternative models or optimizing agent usage to stay within new token-based constraints.

Key insights

Agentic AI workflows significantly increase compute demands, necessitating pricing model adjustments.

Principles

• Agentic AI consumes high compute.
• Per-request billing is unsustainable for agents.

Method

GitHub is transitioning from per-request billing to token-based usage limits per session and weekly to manage the increased compute costs associated with agentic AI workflows.

In practice

• Expect higher costs for agentic AI.
• Monitor token consumption closely.

Topics

• GitHub Copilot
• Pricing Changes
• Usage Limits
• Agentic Workflows
• Compute Demands

Code references

• features/copilot

Best for: CTO, VP of Engineering/Data, Director of AI/ML, Software Engineer, AI Engineer, Tech Journalist

Related on AIssential

• GitHub will start charging Copilot users based on their actual AI usage — AI service providers are shifting to usage-based billing to manage escalating inference costs and ensure sustainability.
• GitHub Copilot Desktop App Targets Parallel Agentic Workflows — The GitHub Copilot app centralizes AI agent management, enabling parallel, isolated development workflows within existing GitHub processes.
• The Hidden Reality of Agentic Coding: Building as a Developer on Free AI Tools in 2026 — Relying on free AI coding tools introduces significant coordination overhead and hidden productivity costs for developers.
• GitHub Just Made a Big AI Coding Move — The new GitHub Copilot app offers model provider choice, unlike Codex, but sign-ups are currently paused.
• 🎙️GitHub’s Mario Rodriguez on AI Coding Agents, Copilot, and the Future of Developers — AI agent quality in late 2025 enabled macro-delegation, accelerating GitHub activity and redefining developer roles.
• The Pulse: AI token spending out of control – what’s next? — Exploding AI agent token spend and vendor demand outstrips budgets, while Meta faces internal morale issues.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Simon Willison's Weblog.