Microsoft, Seeking to Stay Competitive, Slashes Prices for Coding Model

hapabapa via Getty ImagesMicrosoft updated its coding model in a move that appears to position itself as a leading model developer capable of meeting enterprises' business needs and staying competitive in the AI market.On August 11, the tech giant released MAI-Code-1.1-Flash. According to Microsoft, the model produces more efficient code at a quarter of the cost compared to MAI-Code-1-Flash, first released in June.The updated model focuses on CLI tasks (which involve using text or natural language to complete tasks), enabling enterprise developers to use fewer tokens to complete them, according to Microsoft. Previously, users paid $0.75 per million tokens for input and $4.50 per million tokens for output. Now, users pay $0.20 per million tokens for input and $1.20 per million tokens for output on GitHub.Microsoft’s upgrade of MAI-Code is another instance in which an AI model provider is paying attention to enterprises’ concerns about the high cost of using generative AI and creating agentic workflows. Microsoft is not the first to slash prices soon after releasing a new model. Most recently, Google, Anthropic and OpenAI have repriced models at lower price points in response to increased attention to tokenomics, the emerging practice of economizing AI token use.Related:Chinese Company Envisions 100,000 Self-Driving Trucks on Road by 2030Lower Pricing in a Competitive MarketFor Microsoft, though, slashing prices as it did could help it become even more competitive in a landscape where Anthropic dominates as the choice coding model provider for enterprises, along with competition from OpenAI’s Codex and SpaceX’s Cursor. In recent months, Mistral and other AI vendors have also released models heavily focused on coding.“For enterprises, they’ll be looking at the lower token cost of [MAI],” said Lian Jye Su, an analyst at Omdia, a division of Informa TechTarget. “Right now, the token cost is through the roof. Any models that can help them to reduce that and yet deliver a high-quality coding experience or software accuracy are quite important.”He added that it would be interesting to see how 1.1-Flash stacks up against SpaceXAI’s Grok 4.5 in terms of performance, as that model also focuses on coding at a lower price point. Grok 4.5 is significantly more expensive, with input costing $2 per million tokens and output $6 per million tokens.For Microsoft, though, the update to MAI is more than just pricing; it's about Microsoft remaining active in the AI market after ending its exclusivity agreement with model provider OpenAI earlier this year.“It’s about retaining users,” Su said. He added that with the other major cloud providers, AWS and Google, fielding their own models, it’s no surprise that Microsoft will continue to push its own models to enterprise developers.Related:IBM Signs $240M Deal for Nvidia-Powered AI Cluster“It’s about keeping up with competition,” he continued. Moreover, for enterprises already integrated with Microsoft's other products, such as Copilot and GitHub, using any of the MAI models might seem like the natural next step.About the AuthorNews Writer, AI BusinessEsther Shittu has covered AI technologies and industry trends since 2021. As co-host of the Targeting AI podcast, she talks with experts, thought leaders and practitioners exploring critical AI developments. Before AI Business, she wrote for SearchEnterpriseAI, the New York Daily News, Bklyner and the Brooklyn Daily Eagle. When she's not diving deep into the world of AI, she spends her time on passion projects and raising her three daughters.