Open-Source PhotoCraft Emerges as an AI-Coded Challenger to Adobe Photoshop

It was perhaps only a matter of time before developers pointed modern AI coding tools directly at industry behemoths to see what could be built. The result of that inevitable collision is PhotoCraft, an open-source, cross-platform image editor written in Rust that closely mimics the interface, layout, and feature set of Adobe Photoshop.

The application forms part of a broader suite of creative software being developed under the ArtCraft project, led by Brandon Thomas, the former CEO of storyteller.ai. Positioned by its creators as a clean-room reimplementation of Adobe Photoshop, the software has been engineered to avoid potential copyright infringement. Because the developers have carefully avoided reverse-engineering any proprietary, undocumented code, relying instead on publicly available specifications and observed behaviors, the project is already generating significant enthusiasm among free and open-source software (FOSS) advocates.

Despite currently residing in its early alpha stages, PhotoCraft is remarkably well-stocked with features. It includes comprehensive support for layers, masks, layer styles, smart objects, and a fully functional brush engine. Furthermore, the application boasts thirty-four distinct tools, twenty-seven blend modes, and a variety of filters, alongside the crucial ability to open and save files directly in Adobe’s native .psd file format.

In terms of user experience, the software deliberately mirrors the placement of menus, keyboard shortcuts, and toolbars found in the real deal. This design choice is intended to allow professionals and enthusiasts who are already deeply familiar with Photoshop to transition into the open-source alternative without having to relearn basic navigation or muscle memory.

Under the hood, GPU-acceleration is handled natively across multiple operating systems, utilizing Vulkan on Linux, Metal on macOS, and DirectX 12 on Windows, alongside a WebGPU option available through its WebAssembly build for those wishing to self-host. Uniquely tailored for the modern era of software development, the application is also packed with extensive agentic hooks built into virtually every feature. These hooks allow autonomous AI agents to interact with and utilize the application programmatically, reflecting its cutting-edge design origins.

Performance-wise, the application starts up rapidly, proving remarkably responsive even on modest hardware configurations like a standard workhorse laptop equipped with an entry-level processor and integrated graphics. True to its inspiration, it is also capable of consuming significant memory resources shortly after loading complex layered files, mimicking the resource usage of Adobe’s flagship software.

The motivation behind the project stems largely from developer frustrations with the proprietary software ecosystem. Public statements made by Thomas on community platforms indicate that the primary catalyst for cloning Adobe’s software involved navigating the steep subscription fees, administrative hurdles, and general friction associated with attempting to cancel an existing Adobe subscription. For a developer with a deep background in artificial intelligence and software engineering, creating a direct alternative presented an enticing and logical target.

It Won’t Replace Photoshop for Pros

For seasoned users intimately familiar with Adobe Photoshop, approaching a newly minted open-source clone naturally invites a degree of skepticism. Having spent considerable time testing PhotoCraft, the user interface and overall user experience come surprisingly close to the real thing—significantly closer than previous historical efforts aimed at simply skinning traditional open-source editors like GIMP to look like Photoshop.

Instead, PhotoCraft feels spiritually indebted to projects like Photopea, the browser-based, human-coded image editor that has successfully replicated a Photoshop-like environment on the web for many years. However, superficial similarities can be deceptive. Having identical features positioned in familiar locations with corresponding names and icons is fundamentally distinct from those features behaving or rendering output in an identical manner.

While PhotoCraft achieves an uncanny visual resemblance to Photoshop, it remains some distance away from actually feeling and functioning like Photoshop under the hood. Adobe has spent decades fine-tuning its blending modes, complex filters, content-aware tools, advanced object detection algorithms, and spot-removal systems. While open-source clones can study public specifications and emulate observed behaviors, they cannot easily replicate decades of proprietary mathematical fine-tuning and exact output fidelity.

Whether these discrepancies matter depends entirely on the proficiency of the user. Casual individuals performing basic image edits may scarcely notice a difference until they attempt to follow along with a detailed, advanced tutorial on video platforms. For that specific audience, numerous capable alternatives already exist, and even commercial alternatives like Affinity applications can run smoothly on Linux platforms using compatibility layers.

For professional graphic designers and digital artists, however, missing features, subtle bugs, divergent output results, and the accumulation of minor operational differences quickly add up in a high-pressure, day-to-day workflow. Whether PhotoCraft ultimately matures into a fully realized production tool or simply serves as a high-profile technical demonstration—and a statement against prevailing industry pricing models—remains to be seen.

Try PhotoCraft on Ubuntu

PhotoCraft is distributed as free, open-source software with native support for Windows, macOS, and Linux. The project’s official GitHub release page hosts alpha builds tailored for all major operating systems, including a dedicated DEB installer compatible with both Intel/AMD and ARM64 architectures for Ubuntu users, alongside portable AppImage options.

Although the core software application is free and open source, the broader ArtCraft ecosystem operates on a tiered subscription model via its official website. Purchasing a monthly or yearly subscription grants users access to AI credits intended for upcoming generative artificial intelligence features within the ecosystem, though those specific cloud-enabled features are still rolling out across the platform.

Reflecting the rapid pace characteristic of modern AI-assisted software development, the release cycle has been remarkably brisk, with multiple minor and major versions rolling out in rapid succession shortly after initial deployment. This rapid iteration, combined with widespread user fatigue surrounding proprietary software subscription models, has generated massive community interest and goodwill, pushing the project’s repository to accumulate tens of thousands of stars on GitHub in a very short span of time.