Apple has quietly introduced the ninth generation of its in-house RAW processing framework, a significant update that will ship within iOS 27, iPadOS 27, and macOS 27 later this year. The new engine, known as RAW 9, arrives nine years after its predecessor and expands native compatibility to more than 784 camera models. A full list of supported hardware is available on Apple’s website.
Improved noise reduction and detail preservation
The most striking enhancement is an overhauled noise reduction system. During a demonstration using a Canon EOS 5D Mark III frame captured at ISO 51,200, RAW 9 visibly suppressed image noise while retaining considerably more fine detail than the previous engine. Apple also states that the update yields more accurate color rendering and sharper overall output, benefits that should be noticeable across a wide range of third-party apps that rely on Core Image for their editing pipelines.
Machine learning drives a heavier workload
Because RAW 9 depends on machine learning models, it places greater demands on system resources than RAW 8. Apple notes that baseline image processing can be cached, meaning that subsequent edits do not force a full re-run of the engine from scratch. In practical terms, this means performance is expected to remain acceptable even on current hardware such as the iPhone 17.
Developer adoption and professional tooling
Detailed technical information is available through a free developer session. For end users to see the improvements, individual applications must actively support the new engine. Professional desktop tools like Adobe Lightroom and Capture One use their own proprietary RAW decoding pipelines and will therefore not rely on RAW 9. On iPhone and iPad, however, a large number of third-party camera and editing apps depend on Core Image, and those titles should gain access to the improved processing as developers ship updates later this year.
Sources: www.amazon.com, developer.apple.com, 9to5mac.com