mAi 8435817ce1 mAi: #10 - multi-model backend expansion (workflow templates + compare harness)
Path 1 architecture: one comfyui adapter, workflows as data.

- workflow_template.go: embed.FS + token substitution with type-preserving
  whole-value placeholders. ${prompt} → string, ${seed} → int64,
  ${cfg} → float64 — no JSON round-tripping. Partial matches ignored.
- comfyui.go: refactored to load workflow from embedded FS or filesystem
  path. Back-compat preserved: workflow: defaults to flux1-schnell.
- workflows/{flux1-schnell,flux2-klein,sd35-medium}.json — bundled
  templates. flux1-schnell migrated from hardcoded with identical node IDs.
- compare.go: new `imagen compare` subcommand. Sequential N-backend run
  (one GPU on mRock — parallel would OOM), per-backend PNG, sidecar JSON
  with per-model metadata + errors, composite contact sheet via Go image
  package (no ImageMagick dep).
- Sample config gains flux2-klein-local + sd35-medium-local instances.
- docs/backends.md: architecture rationale + per-model HF download paths
  + how to add a new bundled workflow + compare-harness reference.

Live smoke verified: compare mock + flux-schnell-local at 768×768 →
both PNGs written, sidecar JSON has workflow="flux1-schnell" + full
metadata, contact sheet renders. Worker contract (Request → Generate)
unchanged, so flexsiebels /imagine UI API surface preserved.

Tests: 11 existing comfyui + 6 new workflow_template + 5 new compare
tests, all green.

Adding a new model is now yaml + JSON, never Go.
2026-05-11 17:29:57 +02:00
2026-05-08 12:25:21 +00:00

ImaGen

Model-agnostic image-generation framework: pluggable backends (local FLUX on mRock, Replicate, DALL-E, …) behind a single CLI / skill / API.

imagen generate "a cat in a fishbowl" --backend flux-schnell-local --size 1024x1024

See CLAUDE.md for the design — backend contract, registry, config layout, how to add a new adapter.

Install

go install mgit.msbls.de/m/ImaGen/cmd/imagen@latest

Or from a checkout:

make build       # writes ./bin/imagen
make install     # installs into $GOBIN (defaults to ~/go/bin)

First run

mkdir -p ~/.config
imagen config init > ~/.config/imagen.yaml
imagen config validate
imagen backends
imagen generate "test prompt" --backend mock --output /tmp/x.png

The mock backend ships in this repo and produces a deterministic gradient PNG — useful for smoke-testing the pipeline without reaching any model.

Status

Component Status
Backend interface done (#1)
Mock backend done (#1)
ComfyUI / FLUX on mRock open (#2)
Replicate adapter open (#3)
/imagine skill open (#4)
HTTP server stubbed (#1)

Issues live at https://mgit.msbls.de/m/ImaGen/issues.

Description
Model-agnostic image-generation framework: pluggable backends (local FLUX on mRock, Replicate, DALL-E, …) behind a single CLI/API/skill.
Readme MIT 161 KiB
Languages
Go 99.3%
Shell 0.5%
Makefile 0.2%