diffusers-workflow: declarative JSON pipelines on Diffusers, agent-ready via MCP
dkackman11 · reddit · 2026-09-04
Open-source project diffusers-workflow replaces node graphs with a declarative JSON document describing your generation pipeline, running directly against Hugging Face's Diffusers with no abstraction layer. Web UI forms are generated by introspecting real pipeline signatures, catching typos before any model loads.
Highlights:
- Built-in MCP server: models, prompts, workflows, and outputs are all agent-accessible; the author has Claude create prompts and run complex multi-shot workflows from simple ideas, troubleshooting generation issues on its own.
- Multi-step pipelines (text-to-image → image-to-video, inpainting, ControlNet) with data flow and variable substitution.
- Persistent GPU worker/REPL keeps models loaded; supports BitsAndBytes, TorchAO, GGUF, TeaCache, FirstBlockCache quantization/acceleration.
- LoRA, IP-Adapter, prompt library with AI-enhance; CLI, REPL, web UI, and MCP share one API.
- Cross-platform: CUDA, Apple Silicon (MPS), CPU.
Early stage, aimed at terminal/Claude Code/venv users; feedback welcome. Repo: github.com/dkackman/diffusers-workflow
More from coding & agent
- GPT-6 Astra builds explorable 3D city scenes in Unity, 1.9x faster than GPT-5.6 Sol — ChrisGPT · 2026-09-04
- Dev builds a working macOS 27 simulator with Astra in 75 minutes, cloud sync included — pvncher · 2026-09-04
- Opinion: AI Can Amplify Technical Debt — craigmullins · 2026-09-04
- Sandbox readiness is the new bottleneck in agentic apps; git as state primitive — w_hgm · 2026-09-04
- OSWorld Creator: From Operator to GPT-6 Astra on OSWorld 2.0 — TianbaoX · 2026-09-04
- Theo: GPT-6 Astra one-shots a 3D browser game, world-class at Blender and spatial reasoning — charliermarsh · 2026-09-04