AI BriefingGitHub Copilot Sharply Raises Per-Model Usage Multipliers
Key point
GitHub Copilot is raising its per-model usage multipliers as it shifts toward a usage-based billing system.
Details
GitHub recently updated Copilot's per-model Multiplier table, signaling a sharp shift in its cost structure.
Key changes include:
- Claude Sonnet 4.6: 1x → 9x
- Claude Opus 4.6: 3x → 27x
This change suggests that AI companies are finding it increasingly difficult to absorb the compute cost gap generated by existing flat-rate models. In particular, agentic workflows and long-context sessions consume 10 to 100 times more tokens than regular chat, rapidly driving up infrastructure costs.
Starting June 1st, GitHub plans to fully transition to a usage-based billing system. This multiplier increase is a preliminary step ahead of the full billing transition, and in corporate environments without per-model usage management, there is a risk that AI-related costs could significantly exceed expectations.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
US Appeals Court Upholds Pentagon Blacklisting of Anthropic as Supply Chain Risk