Apple’s M5 Ultra Mac Studio Starts at $5,499 as AI API Rates Fall

Apple’s $5,499 Mac Studio with M5 Ultra gives AI users a local-inference option as OpenAI and Anthropic lower token prices, putting hardware costs alongside usage-based cloud fees.

Apple announced the new Mac Studio on Aug. 25, and availability began Sept. 22. The M5 Ultra model starts at $5,499 and is configurable with up to 512GB of unified memory. The 512GB configuration is expected in late October.

The Mac Studio can run large AI models entirely on the device. Four systems can deliver up to three times the AI inference performance of one system, and the 512GB configuration expands the system’s available unified memory for local model use.

A developer demonstration showed LM Studio founder Yagil running Kimi K2.6 across four Mac Studios. The model was described in the demonstration as having one trillion parameters. “There’s no token cost,” Yagil said.

Cloud API prices also changed on Sept. 22. OpenAI listed GPT-6 Sol at $2 per 1 million input tokens and $10 per 1 million output tokens, 50% below GPT-5.6 Sol’s promotional rates. Anthropic set Claude Opus 5.5 at $4 per 1 million input tokens and $20 per 1 million output tokens, 20% below Opus 5.

The prices describe different cost structures. API charges depend on token use, while local inference requires an upfront hardware purchase. The figures do not provide a like-for-like comparison for running the same workload, so they do not establish a break-even point between a Mac Studio and an API.