invoke-ai/InvokeAI

This page introduces an external open-source repository. It is not an HDATF product.

What it does

InvokeAI is an installable visual-generation application for artists and designers, not a skill instruction collection. Users supply prompts, images, and generation settings through a locally hosted web server and React interface. Its canvas supports image refinement, painting within an image, and extending an image beyond its existing edges. Node workflows combine processing steps into reusable generation pipelines, while boards and a gallery organize images and their settings. The backend also exposes queue operations for submitting work, reading progress, and cancellation, providing an interface for separately built automation. The recommended starting point is the official launcher, which manages installation; local generation still requires suitable model files, hardware, and drivers. The inspected source requires Python 3.12, and hardware needs vary with the model, precision, and output size rather than one universal memory figure. Some listed models use external APIs, so local installation does not mean every generation route is offline or independent of provider access. The application license does not replace separate review of model and input-media terms, and supported features have not been exercised in this research.

How it helps ATF

This is a candidate external image-production tool for ATF Works explanatory documents and AX consulting proposal illustrations. Harness could submit a reviewed workflow to the queue interface, retain the job identifier and settings, and attach approved outputs rather than treating a screenshot of progress as completion. Such a proposal would need an adapter for queue status, image retrieval, document storage, and access control; installing the application is not that integration. Evaluation should fix a model and workflow, then check output review, cancellation, memory use, and recovery when a completed image has been deleted. Generated illustrations must remain separate from Company Brain source evidence and from inspection or engineering records. It does not replace technical drawings or certify product accuracy, and no connection or production quality has been verified.

License

Apache-2.0 Permissive, with a patent grant. Commercial use is allowed; keep the notices and state your changes.

More in this category

• storytold/filmcraft
  FilmCraft is a Rust video editor for people who assemble footage, sound and captions on a multitrack timeline. It imports media and interchange documents, applies cuts, grading, effects and audio mixing, and exports videos, audio or project exchanges. The architecture separates codecs, project data, rendering and an engine whose commands are shared by the desktop UI, CLI and MCP agent interface. Integer timeline ticks and project snapshots support precise edit positions and undo, while export jobs expose progress and cancellation. Desktop source builds use Cargo for macOS, Windows and Linux, and optional shared fonts support Japanese text. Although the README still calls the browser frontend a future step in one section, the web application README and web documentation describe an existing WebAssembly build requiring the matching wasm-bindgen tool. That browser path is single-threaded and documents unavailable thread-dependent functions, so it should not be treated as equivalent to the desktop path. The project also reports missing HEVC and AV1 export, absent external effect plugin hosting and less real-media testing on Windows and Linux, while its speed and compatibility figures remain upstream claims.
• storytold/artcraft
  ArtCraft is a desktop workspace for artists and filmmakers who want visual control over AI image and video generation rather than relying only on text prompts. Inputs include prompts, reference images, layered compositions and staged 3D scenes. Canvas tools, character posing and camera placement let users shape a scene before requesting generated images or videos. The documented workflows also include background removal, image-to-mesh conversion and world generation through selected providers. Development documentation identifies a Rust and Tauri desktop application with a React frontend and separate backend services, not a self-contained offline model runner. Stable downloads target Windows and macOS, while source development including Linux requires Rust, Node.js, npm and the Tauri CLI. Model access depends on provider services and account conditions, and the README marks some catalog entries as disabled or unavailable in the desktop app. Its unfinished fair-source license restricts commercial software resale and competing products, so public source access must not be confused with unrestricted open-source reuse.
• storytold/lightcraft
  LightCraft is a Rust photo library and development application for photographers who organize and adjust images locally. It imports JPEG, DNG and several camera RAW formats, keeps originals separate from edits, and exports formats including JPEG, PNG and TIFF. Its engine processes light and color in a wide-gamut floating-point pipeline, with a GPU path and CPU fallback. Albums, ratings, search and an operation log connect photo selection with saved development settings. The desktop interface, CLI and MCP, a protocol for exposing tools to agents, share engine commands for importing, adjusting, rendering and exporting. Source builds use Rust and Cargo for macOS, Windows and Linux, while the browser build additionally needs the WebAssembly target and a matching wasm-bindgen tool. The documentation warns that camera color calibration is incomplete outside DNG, some RAW formats only expose embedded previews, and subject and sky masks use heuristics rather than learned segmentation. Its speed and CPU/GPU agreement figures are maintainer reports, not results independently reproduced here.
• storytold/photocraft
  A native Rust image editor with layers, masks, brushes and layered PSD editing. Exposes its editing engine through a UI, CLI, JSON control channel and MCP server.
• edenfunf/reelmimic
  ReelMimic analyzes a reference video's style and proposes a production plan for user approval. Multiple local Claude Code or Codex agents then create and review shots, with corrections when needed.

Only repositories in the ranked Trendshift lists are included, and the lists are used only to find candidates. We do not copy their ranks. Descriptions, licenses and star counts come from each GitHub repository. The notes are our own reading. We have not tested these projects, and a place on a trending list does not prove quality.