News Summary

• Microsoft has released MAI-Code-1-Flash for general availability to GitHub Copilot Business and Enterprise customers.
• This first-party model is specifically designed for high-volume, iterative coding tasks that require low latency and high speed.
• Billing for this model follows a usage-based system where organizations are charged based on the number of tokens processed.
• Microsoft claims the model can solve complex problems using significantly fewer tokens compared to competing small-tier models like Claude Haiku.
• This pricing structure allows teams to optimize their AI spend by matching specific development tasks to the most cost-effective model.