Back to directory

ima2-gen

A local-first visual generation studio that runs image and video jobs from several providers on a user's own machine, keeping every generation's prompt, timing and settings so results can be branched and reproduced.

ima2-gen

ima2-gen is a local-first visual generation runtime and studio for people and coding agents, offering reproducible image and video workflows across multiple providers. It runs a small local server, reachable from a browser, a Mac app, a CLI or a coding agent, and keeps every generated result together with its prompt, timing and settings.

Results can be organized in a node graph, where a good image is branched in several directions at once without overwriting earlier versions, and a canvas mode supports annotating, erasing, cutting backgrounds and exporting with an alpha channel. It connects to providers including OpenAI, Grok, Gemini, AtlasCloud, MiniMax, NovelAI and registered ComfyUI workflows, sending prompts and reference images only to the provider chosen for that job, with Runway and Higgsfield reached through separate MCP integrations.

It installs with npm, or as a signed and notarized native application for Apple Silicon Macs that bundles its own runtime.

More in Generative media

See all 146
  • Blender Copilot

    github.com

    A Copilot-style AI chat panel inside Blender that runs its agent loop in Blender's own Python process, executes the model's generated code against the live scene, and wraps every conversational turn in a single undo step.

  • A ComfyUI add-on that patches ComfyUI-GGUF so Qwen-Image-2.1 models can run fully as GGUF files, fixing a text-encoder mismatch and missing architecture detection until upstream support arrives.

  • A ComfyUI node that expands a short prompt into a full styled prompt, negative prompt and aspect ratio for Qwen-Image-2.1, using one of 17 photographic styles.